Purpose and Scope

SPATT samplers were developed more than 20 years ago (MacKenzie and others, 2004), and despite their widespread global use for cyanotoxins, there is little specific guidance available to ensure consistent methods and data quality across studies. Although a USGS Techniques and Methods report (Alvarez, 2010) has been developed for two other passive sampling devices, semipermeable membrane devices (SPMD) and Polar Organic Chemical Integrated Samplers (POCIS) and includes best practices for general use of passive samplers, there are important SPATT-specific considerations that are not addressed in that report. Because SPATT samplers are an underdeveloped tool for cyanotoxin monitoring, we currently (2026) lack the data needed to compile a field manual or a USGS Techniques and Methods report. To move toward the goal of generating more consistent SPATT data, this report (1) provides information about how to use SPATT samplers in cyanotoxin research and monitoring studies by describing the lessons learned from 12 studies led by USGS researchers and (2) identifies key knowledge and science gaps that could be filled to improve the use of SPATT samplers and help ensure data are being interpreted consistently and appropriately.

Passive Samplers

Passive sampling devices, like SPATT, accumulate dissolved molecules through diffusion, which is driven by a difference in chemical concentration between the surrounding water and sampler material (Górecki and Namieśnik, 2002; Kot-Wasik and others, 2007). As passive sampling devices, they can be deployed in the environment for an extended period without moving parts or the need for close supervision (Górecki and Namieśnik, 2002). The use of passive samplers is expanding in water-quality studies (Kot-Wasik and others, 2007; Silvani and others, 2017; Godlewska and others, 2021) and cyanotoxin monitoring (Zendong and others, 2016; Roué and others, 2018; Kamali and others, 2022) because they can overcome limitations linked to discrete sampling (Kudela, 2017). Advantages of passive samplers include the ability to integrate the occurrence of dissolved molecules over long periods (from days to weeks to months), capacity to capture episodic or ephemeral events, relative low cost, and ease of use (Kudela, 2017; Roué and others, 2018; Kamali and others, 2022). Passive samplers are particularly suited to flowing and tidal systems, which pose challenges to the interpretation of discrete data because of the rapid movement of large volumes of water.

There are two types of passive samplers typically used in aquatic environments: (1) equilibrium and (2) non-equilibrium or integrative samplers. Equilibrium samplers are designed to allow equilibration between concentrations in the surrounding water and the amount adsorbed to the sampler, whereas integrative samplers are designed to not reach equilibrium with the surrounding water during deployment (Kot-Wasik and others, 2007; Kudela, 2017). Integrative samplers such as SPATT, POCIS, and organic diffusive gradients in thin films (o-DGT; table 1), have a high capacity for collecting the compounds of interest and are effective in environments where concentrations of the target molecule vary through time (Kot-Wasik and others, 2007; Kudela, 2017). Integrative samplers can accumulate and concentrate very low, but ecologically relevant, concentrations of a target compound over a longer deployment period (Vrana and others, 2005; Kot-Wasik and others, 2007), making them especially useful in cyanotoxin research.

Although these three integrative samplers have the ability to accumulate a variety of target molecules in a range of aquatic environments, SPATT samplers have been used most widely to target cyanotoxins, and more broadly, algal toxins in recent decades (MacKenzie and others, 2004; Kudela, 2017; Roué and others, 2018). Publications using o-DGT and POCIS samplers are limited to microcystin and cylindrospermopsin accumulation (table 1). POCIS and o-DGT samplers historically have been used to sample pesticides, pharmaceuticals, and metals (Kot-Wasik and others, 2007; D’Angelo, 2019); however, recent studies have examined their utility as microcystin and cylindrospermopsin monitoring devices (Nyoni and others, 2017; Brophy and others, 2019; Yao and others, 2019; Kim and others, 2021; Wang and others, 2022). POCIS samplers, like SPATT samplers, can provide semiquantitative cyanotoxin concentrations. Because o-DGT sampler construction requires a diffusive layer that may or may not be protected by an additional protective layer, they can be calibrated before use in the field to provide quantitative cyanotoxin concentrations (Kot-Wasik and others, 2007; D’Angelo, 2019; Wang and others, 2022). Compared to SPATT or POCIS samplers, o-DGT samplers generally are more difficult to construct (table 1).

U.S. Geological Survey Solid Phase Adsorption Toxin Tracking Studies

There have been at least 12 USGS efforts across 20 states that have included the use of SPATT samplers for cyanotoxin monitoring, research, and outreach (fig. 1; table 2). Collectively, these studies included a broad range of aquatic environments, and their designs and approaches served as the foundation for the synthesis provided in this report. Together, these 12 USGS studies used SPATT samplers to address broad purposes, including (1) assessment of methods and performance for cyanotoxin measurements; (2) understanding how SPATT approaches fit into the cyanotoxin sampling and monitoring toolbox; and (3) evaluation of occurrence, spatiotemporal variability, and transport of cyanotoxins (table 2).

SPATT samplers have been used at the watershed scale to evaluate the spatiotemporal occurrence of cyanotoxins in the Raritan River Basin and Salem River Basin in New Jersey (table 2; accompanying data releases Spitz and others, 2025; Trevino and others, 2025) and numerous river basins in Oregon (Clackamas, North Santiam, McKenzie, Willamette, and Tualatin River Basins [not shown]; Harmful Algal Blooms and Drinking Water in Oregon [ODW] study; fig. 1; accompanying data release Carpenter and Wise, 2023; Carpenter and others, 2025). In these studies, SPATT samplers were concurrently deployed in multiple locations along river corridors to generate a consistent time series of time-integrated cyanotoxin concentrations. In the Harmful Algal Bloom Monitoring in the Finger Lakes Region, New York (FLX) study in New York (accompanying data release Stouder and others, 2024), SPATT samplers were deployed at multiple water depths to study cyanotoxin occurrence throughout the water column. SPATT samplers have also been used at the regional scale to compare cyanotoxin occurrence and diversity among watersheds in Wisconsin, North Dakota, and Minnesota (Beyond Microcystins [BMC] study). Although the regional scale SPATT deployment necessitated lower spatial sampling density compared to a watershed-scale SPATT study, the regional scale can give information on cyanotoxin patterns across ecosystems, climate regimes, and large geological features. A joint USGS and National Park Service study (Rapid Response Strategy for Potential Toxin Exposures from Harmful Algal Blooms in Coastal and Shoreline Areas of National Parks [NPS]) used SPATT samplers to monitor cyanotoxins and establish rapid response protocols in 22 national parks and National Parks Service managed areas across the United States.

SPATT samplers have been used to study the transport of cyanotoxins across the freshwater to marine interface (Cyanobacteria, Cyanotoxins, and Associated Characteristics Across the North Atlantic Appalachian Region [NAR; accompanying data release Trevino and others, 2024], Monitoring Harmful Algal Blooms in a Coastal System to Identify the Factors that Affect HAB Production and the Downstream Transport of Cyanobacteria and Associated Cyanotoxins from Freshwater to Marine Environments [NJC], Monitoring Cyanotoxins in California's Sacramento–San Joaquin Delta: Fixed Stations [SSD; accompanying data release Bouma-Gregson and others, 2025], and Monitoring Cyanotoxins in California's Sacramento–San Joaquin Delta: High-Resolution Mapping Surveys [HSM] studies; table 2). These studies focused on lowland rivers and estuaries to evaluate the transport of cyanotoxins into the coastal ocean. In the Sacramento–San Joaquin Delta, in California (HSM study), SPATT samplers were deployed in a boat-based flow-through chamber to collect spatially integrated data. SPATT samplers were swapped out as the boat traveled through different regions of the Sacramento–San Joaquin Delta, providing insight into the drivers, sources, and transport of cyanotoxins. SPATT samplers have also been used to examine the transfer of cyanotoxins to shellfish in rivers of Tennessee and Virginia (Influence of Harmful Algal Blooms on Mussels in the Clinch River, Tennessee and Virginia [CRB] study). There are many endangered or threatened species of mussels in the southeastern United States, and understanding the effect of cyanotoxins on their health is important to their continued management. Additionally, SPATT samplers have been used as educational tools, helping students from diverse backgrounds learn about environmental biology, water chemistry, laboratory methods, and biotechnology, all while doing cyanotoxin research in local urban waterways (HAB Education and Outreach in Urban Settings, Tennessee [EDU] study). The diversity in geography and scope of these 12 USGS SPATT studies provide numerous lessons learned in the generation and interpretation of cyanotoxin data.

Solid Phase Adsorption Toxin Tracking Sampler Construction

SPATT samplers generally are constructed using mesh and an adsorbent resin to which dissolved cyanotoxins in the environment can adsorb to. However, there are numerous commercially available sorbents and standard sampler designs that can be used depending on the cyanotoxin(s) of interest, deployment environment, and supplies available to a research program. These construction options are described in more detail in the next sections, with examples from the 12 USGS studies listed in table 3.

Sorbents

Numerous studies have assessed the utility of different sorbent resins with a variety of target cyanotoxins (as reviewed by Roué and others [2018]). More than 15 commercially available resins have been used to successfully sorb cyanotoxins using SPATT samplers (Roué and others, 2018). HP20 (DIAION, Mitsubishi Chemical Company, Tokyo, Japan), a porous, cross-linked, styrene-divinylbenzene resin, generally is considered a versatile resin that can adsorb multiple types of cyanotoxins and is commonly used in SPATT samplers (MacKenzie and others, 2004; Kudela, 2017, 2020; Roué and others, 2018). Because HP20 adsorbs a variety of cyanotoxins, it is an especially useful resin to use when multiple cyanotoxins are of interest (Kudela, 2017; Bouma-Gregson and others, 2018; Roué and others, 2018). Kudela (2020) and Negrey and others (2023) note that HP20 offers excellent adsorption and desorption capabilities for numerous cyanotoxins, while other resins like Oasis Hydrophilic-Lipophilic Balance (HLB; Waters Corporation, Milford, Massachusetts), Strata-X (Phenomenex, Torrance, California), and SP700 (SEPABEADS, Mitsubishi Chemical Company, Tokyo, Japan) have different properties that could be appropriate for different toxins, environments, or research objectives. HLB resin has been used to examine nodularins and anatoxins in freshwater reservoirs and lakes and has demonstrated high adsorption, as did HP20 when samplers were co-deployed (Kohoutek and others, 2008; Kudela, 2020; North Coast Regional Water Quality Control Board, 2022b). Strata-X had high capacity and fast kinetics when used to examine microcystins (Kudela, 2020) and has been determined to effectively adsorb anatoxin-a, especially during short deployment periods (Wood and others, 2011). Similarly, SP700 has indicated high adsorption capacities for microcystins in laboratory trials (MacKenzie, 2010; Zhao and others, 2013). The cost of different resins is often a consideration in the design of SPATT samplers. Kudela (2020) noted that the relatively low cost of HP20 has contributed to its common use in SPATT studies.

A range of resin masses have been examined in the literature, but Zendong and others (2016) detected that 3 grams (g) of resin performed better than 0.3 g or 10 g. The USGS studies referenced in this report (tables 2, 3) used 3 g of the HP20 resin as the sorbent, like most researchers deploying SPATT samplers (Kudela, 2017). The HP20 resin may arrive from the manufacturer containing moisture or sorbed water from the atmosphere; if this water is not removed in a desiccator, total resin mass used for a SPATT sampler may not be accurate. Although humidity can be quantified, it is often preferred to remove this water before weighing the resin to ensure the desired mass is consistently used in each SPATT sampler. In preparation for SPATT construction in the HSD and SSD studies (table 2), researchers dried the resin at 60 degrees Celsius (°C) for as much as 72 hours to remove moisture. The resin was then stored in a sealed amber glass jar or covered beaker in a desiccator until use. When weighing resin, a static deionizer can be used to eliminate or reduce electrostatic charges, making the resin easier to weigh and work with. Researchers preparing resin for use in the ODW study (table 2) did not bake resin but instead placed the resin in a desiccator and only proceeded with SPATT construction when the resin readily flowed and no longer visibly clumped together. None of the other USGS studies desiccated or baked the resin before SPATT construction. However, potential issues were noted in the NPS and RRB studies where post-deployment weights were less than the original pre-deployment weights. With no visible loss of resin during the field deployment, and best efforts to collect all resin from the samplers before extraction, researchers hypothesized that the original resin weight may have contained moisture resulting in the higher pre-deployment weight. In these instances, the original pre-deployment resin weight (3 g) is typically factored into the final concentration calculations (Negrey and others, 2023).

The resin needs to be activated before use to remove impurities and swell the resin beads, which opens up the pores and expands the surface area of the resin (Qi and others, 2014). The USGS studies used 100 percent methanol to activate SPATT resin (Lane and others, 2010; Kudela, 2017). Negrey and others (2023) recommend a 24-hour activation period during which the resin is fully immersed in methanol. However, suggested activation periods in the literature range from 2 hours (Li and others, 2016) to 48 hours (Fux and others, 2008). Lane and others (2010) also recommend sonication after methanol rinsing to remove residue, but this step was not incorporated into the USGS studies. After activation, the resin is rinsed with deionized or organic free water and then stored in deionized water at 4–6 °C until deployment (Negrey and others, 2023) because the resin needs to remain hydrated after activation so that the pores do not collapse. All of the USGS studies kept activated SPATT samplers in sealed bags of deionized or organic free water after activation through deployment in the field.

The sorption of cyanotoxins to SPATT resin is influenced by several environmental factors that can introduce potential uncertainty. For example, SPATT samplers bind cyanotoxins, along with a host of other non-target analytes, such as dissolved organic matter (Wu and others, 2011) or salinity (Fan and others, 2014) that may cause interference during deployment or analysis. Armstrong and others (2025) examined how phenolic compounds that can readily bind to SPATT resin may interfere with cyanotoxin adsorption in an estuary. The presence of the phenolic compounds caused matrix interference and greatly reduced microcystin detection (Armstrong and others, 2025). Because cyanotoxins diffuse from the surrounding ambient water through the boundary layer or region of laminar, rather than turbulent, flow and are then sorbed (Booij and others, 2017) to the SPATT resin, changes in the hydrodynamics (water velocity) of a sampling area may impact cyanotoxin sorption. Sorption rates often increase with elevated flow rates, but the direction of flow relative to the SPATT sampler position and turbulence levels may also play a role (Booij and others, 2017) and warrant further investigation.

Sampler Types

There are two common styles of SPATT samplers: (1) the sachet (fig. 2A) and (2) the hoop (fig. 2B). Both styles can be constructed in a variety of sizes and shapes. The sachet style consists of resin encased in a piece of 100-micrometer (µm) mesh nylon fabric, or Nitex (Wildco, Yulee, Florida), that is folded in half and heat-sealed on the three open sides to contain the resin (fig. 2A). The hoop style consists of resin encased between a single piece folded in half, or two separate pieces, of 100-µm mesh nylon fabric, which are then held together by an embroidery hoop (fig. 2B). Alternatively, it is possible to use a combination of both types of samplers, as demonstrated in the HSD and SSD studies (table 2; fig. 2C). This hybrid design features one 12.7-centimeter (cm) hoop with two triangle-shaped, heat-sealed sachets held in place by the hoop (fig. 2C). In these examples, the hoop acts solely as an attachment point and does not seal the nylon mesh. The style of SPATT sampler used, as well as the size of the sachet or hoop, may affect the ability of the resin to adsorb the target cyanotoxin in each sampling medium; however, effects of sampler design on performance have not been studied.

Each SPATT sampler type offers advantages and disadvantages in ease of construction, deployment, and extraction. The sachet generally is lower cost because they are smaller and require less nylon mesh fabric and methanol during activation. The sachet may allow for larger batch sampler preparation because more of the smaller, malleable sachets can be activated at once, as compared to the bulkier, rigid hoops. However, because of their flexible nature, the sachet style sampler has the potential to fold over during a deployment period, which could potentially result in reduced exposure surface area. One key advantage of the hoop is the ease of deployment; the hoop offers a secure attachment point for zip ties or other fasteners, and they do not need a heat-seal, relieving any concern that the heat-seal may fail during deployment. The RRB study (table 2) lost about 5 percent of the sachets deployed because of sealing issues. During the ODW study (table 2), the use of heat-sealed sachets was attempted, but it was discovered that the nylon mesh was prone to melting and, in the end, the deployment of hoops was done to ensure the resin remained secure during deployment. With the hybrid hoop-and-sachet design, despite the additional support from the hoop, sachets were detected during the SSD study without resin on two occasions, suggesting the heat-seal failed. In addition to secure seals, hoops spread the resin across the diameter of the hoop, resulting in a larger surface area in direct contact with the water, which may allow for increased adsorption of the target dissolved molecule, though this needs to be explored further. Sewing in additional patterns to help evenly distribute the resin throughout the surface of the sampler could aid in even exposure of the resin to the cyanotoxins in the water. Hoops and sachet SPATT samplers have both been successfully used across a variety of aquatic systems worldwide, and the decision to implement one style or another will ultimately depend on the study objectives, resources available, and the deployment environment.

Deployment Infrastructure

Depending on site characteristics, such as water depth, site access, and existing infrastructure, there are a variety of ways to deploy SPATT samplers (fig. 3). Typically, SPATT samplers are secured with rope or zip ties to stable infrastructure like a dock, pipe, or post driven into the substrate (fig. 3A). In turbulent conditions, use of a weighted line may be beneficial to ensure the samplers remain submerged at the desired depth (fig. 3B). For further protection, samplers can be enclosed in a wire cage (fig. 3C) or a plastic tube mounted to a secure dock or post (fig. 3D). Such housings may reduce the buildup of debris around the samplers and can prevent loss if, for example, a sampler becomes detached but remains inside the housing. The SPATT hoops deployed in the HSM study (table 2) were mounted inside of a cooler attached to a flow-through system on a boat (fig. 3E). As the boat proceeded through the sampling transects, hoops were replaced every 30–60 minutes, and the volume of water passing through the hoop was measured using a flow meter (table 3). To target benthic cyanotoxins, MacKenzie and others (2011) successfully placed SPATT sachets in boxes mounted just above the sediment surface. The boxes reportedly protected the sachets from wind and sunlight exposure during low tide (MacKenzie and others, 2011). Depending on site characteristics and study objectives, it may be advantageous to deploy multiple SPATT samplers per sampling location. If the nylon membrane of a SPATT sampler tears, or if a SPATT sampler is lost, stolen, or vandalized, a duplicate sampler could be useful. In the ODW study (table 2), researchers placed SPATT samplers in two locations to help ensure a sampler would be present upon retrieval at priority sampling areas.

Deployment Length

The SPATT sampler deployment length will vary and depends on study objectives, the environment, and the cyanotoxins targeted. Because cyanotoxins may begin to desorb from the resin after a period of time because of approaching equilibrium, uncertainty of SPATT results often increases as the deployment duration increases. Deployments ranging from minutes to hours will be more kinetic, with more adsorption than desorption occurring (Kudela, 2017). As deployment periods lengthen from days to weeks, SPATT samplers behave more like an equilibrium sampler because cyanotoxins will desorb from the resin as environmental concentrations change (Kudela, 2017). If saxitoxin is being monitored, Rodríguez and others (2011) suggest limiting the deployment length to no more than 7 days to limit the loss of toxin from the sampler because saxitoxin is extremely hydrophilic and tends to “leak” out of synthetic resins. Studies that included the examination of saxitoxin with SPATTs have shown mixed results (Lane and others, 2010; Hattenrath-Lehmann and others, 2018). In addition, the potential for interference from fouling by suspended particles, periphyton, or animals may increase with longer deployments. Such fouling not only potentially affects adsorption and desorption kinetics, but the presence of organic and inorganic particles can complicate extraction procedures by clogging filters, which was the case during the RRB study (table 2) where sediments clogged the filters, adding substantial time to the extraction process. A solution recently implemented by researchers at the USGS Upper Midwest and New York Water Science Centers to avoid clogging is the addition of quartz wool to the extraction column. The wool is added before the resin and included in all pre-rinse steps. The quartz wool helps create a second filter barrier and has shown promise in reducing clogging in the extraction column.

As deployment lengths increase and equilibrium is approached or achieved, cyanotoxins may desorb from the SPATT resin, resulting in a loss of information from the start of the deployment period (Kudela, 2020). The time-weighted average concentrations are dependent on the properties of the sampler (resin), mass of resin, water conditions, chemical properties of the cyanotoxin(s), and exposure period (Górecki and Namieśnik, 2002; Kudela, 2011). When studying microcystin kinetics in a simulated field study, Kudela (2020) determined that the HP20 resin provides a representative time-weighted average after 7 deployment days, and anything longer is weighted toward the end of the deployment period. Therefore, deployments longer than 1 week would be biased toward the final days of the deployment period. Effects of deployment length on the SPATT results were evaluated by the California North Coast Regional Water Quality Control Board (2022a) and the SRB study by deploying multiple samplers at once and retrieving them at different times (Box 1—Assessing the Impact of Deployment Length on Solid Phase Adsorption Toxin Tracking Concentrations). These studies show that longer deployment times do not consistently result in higher cyanotoxin concentrations extracted from SPATT samplers. Additional testing in the ODW study (table 1) indicated consistently lower concentrations with long deployments (26–34 days) compared to shorter, weeklong deployments. Oregon drinking water researchers hypothesize that the lower concentrations obtained from the month-long deployments could be the result of equilibrium changes through the sampling period or possible degradation or desorption of the cyanotoxins. Although the ability to integrate averages through time makes it tempting to deploy SPATT samplers for multiple weeks, longer deployments do have some drawbacks, and researchers must balance their tolerance for uncertainty in the SPATT result, expected variation of cyanotoxins in the environment, and resources available to support sampling efforts when deciding the deployment length for a project.

SPATT sampler deployments in most published studies have been from days to weeks (Roué and others, 2018; Anderson and others, 2023). Most of the SPATT sampler deployments for the USGS studies listed in table 2 were done for several weeks (table 3). The SPATT sampler deployment frequency often matches the sampling frequency for many discrete water-quality monitoring programs, which enables field staff to collect samples on a consistent day of the week. Shorter deployments also have been done with success. Wood and others (2018) did diel SPATT sampling with sequential 2-hour deployments to track sub-daily variations in anatoxin concentrations. Additionally, the HSM study (table 3) and Peacock and others (2018) deployed SPATT samplers in flow-through systems for 30-plus minutes to track spatial variation in cyanotoxin concentrations along a boat track. When collecting discrete samples from a site, multiple-hour SPATT deployment can be convenient if other samples or data are collected at a site for a similar period of several hours. A SPATT sampler can be deployed upon arrival at the site and subsequently retrieved when all other sampling is completed.

Fouling

Fouling, the accumulation of organic, inorganic, or biological material on instrumentation submerged in water, often affects sensor operation and data quality (Vanysacker and others, 2014; Delgado and others, 2021). On passive samplers like SPATT hoops or sachets, fouling has been known to affect permeability (Richardson and others, 2002; Lane and others, 2010), which could lead to over or underestimates of cyanotoxin concentrations. Fouling reduces flow through the sampler and limits contact between the resin and the water column, potentially leading to underestimates of cyanotoxin concentration. An additional fouling consideration is the contamination of SPATT resin by particles that would increase the final dry weight of the resin, which is used to normalize cyanotoxin concentration during the final calculation steps (Negrey and others, 2023). In these instances, the ODW study (table 2) included using the pre-deployment resin weight during the final calculations to ensure any additional, unwanted weight did not underestimate cyanotoxin concentrations. The effects of fouling on sample mass can be reduced by thoroughly rinsing SPATT samplers with field water during retrieval to wash away fine particles (Negrey and others, 2023), although this may result in the loss of saxitoxin from the sampler because of its hydrophilic nature; if saxitoxin is of interest, these rinses should be considered as part of the extract (Rodríguez and others, 2011). Other common anti-fouling strategies used for aquatic devices deployed for periods of time include the use of specialized coatings or paint and wipers or brushes (Delgado and others, 2021). However, currently (2026) these anti-fouling approaches have not been tested in conjunction with SPATT samplers.

Biofouling, or the growth and deposition of bacterial cells on a membrane (Vanysacker and others, 2014), may affect SPATT samplers but has not been the focus of research studies to date (2026). Copper cages or mesh envelopes have been successfully used with other passive samplers (Jeong and others, 2018) to deter biofouling (Magin and others, 2010), but these techniques have yet to be implemented with SPATT samplers. USGS researchers have hypothesized potential positive and negative feedback scenarios that could arise from biofouling. For example, if toxigenic cyanobacteria are able to grow on SPATT sampler surfaces (or on cages or pipes if enclosed), locally produced cyanotoxins could also be adsorbed onto the resin, resulting in overestimates of ambient cyanotoxin concentrations. Conversely, bacteria capable of cyanotoxin biodegradation (Edwards and others, 2008; Chaffin and others, 2022) could grow on SPATT sampler surfaces, potentially degrading cyanotoxins before they can be adsorbed onto the resin. In this scenario, the cyanotoxin concentration derived from the SPATT sampler may underestimate ambient cyanotoxin concentrations. Because expression of biodegradation genes has been linked to previous microcystin exposure (Dexter and others, 2021), this scenario would be most relevant to water bodies with recurring cyanobacteria blooms. Wood and others (2011) suggest that future SPATT experiments could examine the ability of various membranes to prevent biodegradation once cyanotoxins are sorbed onto the resin.

Deployment Depth

Like deployment length, appropriate deployment depths for SPATT samplers will be site- and study-specific. Deployment depth could target a known depth of interest, such as drinking water intakes (Beecher, 2022) or bivalve habitats (MacKenzie and others, 2011). Additionally, if historical phytoplankton or environmental data were collected at fixed depths, SPATT sampling efforts could also be focused there. Alternatively, multiple SPATT samplers could be deployed to capture differences in concentration through the water column, which may be informative when studying stratified systems. During the FLX study (table 2), this approach was used to examine cyanotoxin concentrations at near surface (1 m), mid depth (13 m), and bottom depth (26 m) in several New York lakes (Box 2—Solid Phase Adsorption Toxin Tracking Use to Determine Cyanotoxin Occurrence Across Multiple Depths). In this study, increased microcystin concentrations at bottom depths later in the sampling season were detected, which would have been missed if only surface samples were considered (fig. 5). During the ODW study, SPATT samplers were deployed at varying depths, at the surface and at 1 m deep. After 1-month deployments, cylindrospermopsin and anatoxin concentrations were elevated at the 1 m depth. Anatoxin concentrations were over 30 times higher at 1 m compared to surface samples, and cylindrospermopsins were above the detection limit only at 1 m depth (Carpenter and Wise, 2023). These studies highlight the importance of considering multiple deployment depths, as ecologically relevant increases in cyanotoxin concentration may occur deeper in the water column across varying levels of temperature, viscosity, or other parameters.

When deploying SPATT samplers at sites with variable water elevations, caution must be taken to ensure that the SPATT samplers remain submerged. The NAR study (table 2), for example, lost numerous sachet style samplers because water levels dropped during the deployment period. Floats can be used to keep the samplers at a desired depth despite changes in water elevation (fig. 3C). In macrotidal systems prone to large changes in water elevation, the use of weighted lines (fig. 3B) or securing SPATT samplers to submerged structures (fig. 3A) may ensure the samplers remain submerged. Deployment depth also may reduce effects of physical damage in high energy aquatic environments or interference by humans through vandalism or theft. During the FLX study (table 2), it was reported that SPATT samplers deployed near the surface were subject to enhanced loss or damage compared to samplers at mid or near bottom depths (Johnston and others, 2025). About 78 percent of SPATT samplers deployed at the surface were recovered intact, whereas more than 90 percent of the samplers deployed deeper into the water column were successfully retrieved. Deploying SPATT samplers with a tag listing the contact information of the study lead and a brief description of the sampler purpose may help prevent vandalism if the samplers are encountered by the public (fig. 6). It may be good practice to deploy samplers in protected areas that are inaccessible to the public or at deeper depths to minimize effects of wave or current action and theft or damage. It also might be helpful to deploy multiple SPATT samplers at a site as a precaution in the case of loss, damage, or theft.

Deploying SPATT samplers at depth or in shade may deter effects of photodegradation on cyanotoxins sorbed to SPATT samplers. The transparency of the nylon mesh means any sorbed organic compounds are potentially susceptible to photodegradation. Numerous cyanotoxins have been shown to rapidly degrade upon exposure to different types of radiation: photosynthetic active radiation, ultraviolet A (UV-A), and ultraviolet B (UV-B; Wörmer and others, 2010; Kurtz and others, 2021). There are still large data gaps and uncertainty of the degradation rates and transformation products for other cyanotoxins, such as anatoxins, nodularins, and variants of cylindrospermopsin and microcystin. In lentic systems, like the FLX study (Box 2—Solid Phase Adsorption Toxin Tracking Use to Determine Cyanotoxin Occurrence Across Multiple Depths), epilimnion turnover and depth distribution of SPATT samplers with regards to light availability must be considered. The effect of photodegradation on SPATT sampler concentrations is yet uncertain and warrants further investigation.

Extraction Methods

There is not a standard USGS method for the extraction of cyanotoxins from SPATT samplers. The most common methods used in USGS studies are based on the Negrey and others (2023) method, adapted from Lane and others (2010) and Kudela (2011). Upon retrieval, SPATT samplers are lightly rinsed with ambient water and shaken to remove detritus, sediment, and any excess water. Removing particles that have fouled the SPATT sampler, especially sediment, helps ease the extraction process by reducing the chance of filter clogging during extraction. After rinsing, SPATT samplers are stored on wet or dry ice during transport and either immediately extracted or kept frozen until extraction in a laboratory. Although SPATT extracts typically are kept frozen (less than or equal to 20 °C) before and in between individual analyses, it is unknown how fast the cyanotoxins in the extract degrade when frozen. The degradation rate likely is dependent on the individual cyanotoxin, the final solvent concentration (50-percent methanol or reconstituted in lower concentration of methanol depending on the cyanotoxin and analysis being used), as well as the number of times (and duration) the extract is removed from frozen storage for analysis. For microcystins, anatoxin-a, and nodularin, Negrey and others (2023) reported no difference in recovery of each toxin after 3 months in storage at −20 and −80 °C.

Cyanotoxin extraction from the SPATT resin is done using a known volume of organic solvent to desorb organic compounds from the resin (MacKenzie and others, 2004; Lane and others, 2010; Zendong and others, 2016). In the Negrey and others (2023) method, SPATT samplers are disassembled and transferred to chromatography columns. The cyanotoxins are then eluted from the resin by pulling a solvent through the column under light vacuum pressure. The method uses three extractions of either 50-percent methanol, 50-percent methanol with ammonium acetate, or 100-percent methanol with formic acid, depending on the cyanotoxin to be analyzed (table 2 in Negrey and others, 2023). The extracts can be combined before analysis or analyzed individually and then summed together (Negrey and others, 2023). Although the Negrey and others (2023) method has been widely used, there are several other published methods for extracting cyanotoxins from SPATT samplers (MacKenzie and others, 2004; Fux and others, 2008; Wood and others, 2011, 2012; Zhao and others, 2013; Zendong and others, 2016; Hattenrath-Lehmann and others, 2018; Onofrio and others, 2020). Additionally, there are methods similar to Negrey and others (2023), which either involve soaking the resin in methanol, followed by filtration, drying and reconstitution (MacKenzie and others, 2004; Hattenrath-Lehmann and others, 2018), placing SPATT resin into an extraction column and eluting cyanotoxins with solvent under gravity or a vacuum (Fux and others, 2008; Zendong and others, 2016), or placing resin into a container, submerging with a solvent, and then shaking for a period of time (Wood and others, 2011, 2012; Zhao and others, 2013). After extraction, an aliquot of the sample is then collected and the cyanotoxins are analyzed, typically using ELISA or liquid chromatography and mass spectrometry (LC-MS) methods.

Although a 50-percent methanol solution has been used as the extraction solvent for SPATT work in USGS and other studies (Negrey and others, 2023), other extraction solvents could yield better recoveries for all or individual cyanotoxins. For example, Negrey and others (2023) suggests using 100-percent methanol combined with 2-percent formic acid in the first extraction step to improve recovery for anatoxin-a because it is notoriously unstable in alkaline conditions. Additionally, analysts in the SSD study (table 2) determined that using formic acid in conjunction with methanol improved recovery for numerous microcystin congeners (fig. 7). Otherwise, little work has been done to investigate improved extraction methods with different cyanotoxin and resin combinations. Another source of uncertainty is if the cyanotoxins are fully desorbed from the resin during the extraction procedure or if some remain behind, possibly resulting in a low bias in the final cyanotoxin concentration measured in the extract. The desorption properties of each cyanotoxin may be affected by other compounds, organic matter for example (Wu and others, 2011), that may bind to the resin during deployment, making desorption characteristics somewhat dependent on the deployment site.

Analyzing Solid Phase Adsorption Toxin Tracking Sampler Extracts—Enzyme-Linked Immunosorbent Assay

ELISAs are widely used to indirectly examine cyanotoxin presence and concentration (He and others, 2016; Filatova and others, 2020). ELISA methods are relatively inexpensive, can be learned quickly by most laboratories, and are commercially available. Several ELISA kits are produced by several manufacturers for measuring microcystins and nodularins, anatoxins, saxitoxins, and cylindrospermopsins. ELISA methods are particularly beneficial as a screening tool to determine if cyanotoxins are in a sample and to track relative changes in cyanotoxin concentration through time (Guo and others, 2017). Despite these advantages, ELISA has seen little use as an analytical methodology for SPATT extracts, other than its use by MacKenzie and others (2004), Lane and others (2010), and Kudela (2011); of these, only Kudela (2011) studied a freshwater system. Additionally, the FLX study compared ELISA with MS methods for analyzing SPATT extracts (Johnston and others, 2025).

There are several disadvantages to using ELISA methods to analyze environmental samples. ELISA methods cannot differentiate among various congeners of a specific cyanotoxin in a sample, so ELISA results often are reported as total cyanotoxin concentrations (North Coast Regional Water Quality Control Board, 2022a). Additionally, because of the structural similarities between microcystin and nodularin molecules (both contain the ADDA [2S,3S,4E,6E,8S,9S)-3-Amino-9-methoxy-2,6,8-trimethyl-10-phenyl-4,6-decadienoic acid] moiety), these cyanotoxins cannot be differentiated using ELISA, and results typically provide the sum of microcystins and nodularins in a single sample (Fischer and others, 2001). The ELISA method also has a relatively narrow dynamic range without prior dilution, and cyanotoxin concentrations cannot exceed about 5 micrograms per liter (µg/L) depending on the specific ELISA kit used (Guo and others, 2017). Because SPATT samplers have a high capacity for concentrating cyanotoxins, it is common for the extracts to exceed this maximum allowable concentration for the ELISA analysis. If this exceedance occurs, samples must be diluted and reanalyzed to bring the concentration within the ELISA test range, requiring additional expenses in labor and supplies. An important consideration noted by Jia and others (2022) is that, depending on the dilution factor used, cyanotoxin concentrations can differ, sometimes by a factor of more than 2. This difference is because the calibration curve is most accurate at and near its inflection point where the ratio between signal change and concentration is greatest (Sasaki and Mitchell, 2001). As such, the accuracy of a given dilution may depend on where on the calibration curve it falls. In addition, if the salinity of the final extract is not known, the accuracy of the quantitation may be affected because many commercial ELISA kits will not function properly outside of their specified salinity range (Gold Standard Diagnostics, 2024a, b, c, d383940); this may present a problem when analyzing estuarine samples.

Methanol, the recommended SPATT extraction solvent (Negrey and others, 2023), also can interfere with ELISA chemistry (Metcalf and others, 2000). Metcalf and others (2000) determined that methanol can mobilize key components of the ELISA microplate wells, resulting in an overabundance of binding sites and overestimates of cyanotoxin concentrations. Diluting SPATT extracts that are composed of 50–100 percent methanol is required to avoid these possible matrix interferences. Alternatively, extracts can be evaporated to dryness and reconstituted in an aqueous matrix with a methanol concentration below the maximum specified by the kit manufacturer (Gold Standard Diagnostics, 2024a, b, c, d). This evaporation step can be done at room temperature in a fume hood or using a nitrogen evaporator. The CRB and EDU studies (table 2) used an evaporator manifold to reduce methanol concentrations and then reconstituted each sample with high quality, distilled laboratory water. It is important to note that this method potentially concentrates any contaminants present that may inhibit ELISA analysis. An important consideration if analyzing multiple cyanotoxins with ELISA is to ensure that the methanol concentration is below the lowest methanol limit across all assays.

Given the interferences previously noted with immunoassays, such as organic matter, salinity, and phenolic compounds (Wu and others, 2011; Fan and others, 2014; Armstrong and others, 2025), along with the potential false negatives or positives and high bias demonstrated by the FLX study (Johnston and others, 2025), there is still uncertainty regarding the effectiveness of ELISA as an analytical method with the SPATT extract matrix. With the experience of other researchers in modifying immunoassays to eliminate matrix interference, for example by adding methanol to standards (Nunes and others, 1998; Huang and Sedlak, 2001; Hanselman and others, 2004; Silva and others, 2014; Liu and others, 2023), it is likely that these matrix issues can be resolved with further laboratory experimentation.

Analyzing Solid Phase Adsorption Toxin Tracking Sampler Extracts—Mass Spectrometry

Mass spectrometry (MS) methods are widely used to directly quantify cyanotoxin concentration and composition with high sensitivity and high throughput (Humpage and others, 2010). Notably, MS methods can measure concentrations of individual cyanotoxin congeners (Moore and others, 2016), but quantification is restricted to congeners that have analytical standards available (Jia and others, 2022). Given that only a few dozen known congeners have standards, MS methods can underestimate cyanotoxin concentrations if a sample is dominated by a congener for which no standard has been developed (Birbeck and others, 2019; Jia and others, 2022). Unlike the ELISA methods, MS requires a skilled analyst and expensive instrumentation; however, once MS methods are established and validated, sample analysis can be time and cost effective.

Comparison studies have demonstrated discrepancies between ELISA and MS cyanotoxin results (Birbeck and others, 2019; Jia and others, 2022; Box 3—Analysis of Solid Phase Adsorption Toxin Tracking Extracts by Enzyme-linked Immunosorbent Assay and Mass Spectrometry Methods). Cyanotoxin concentrations obtained through MS methods are often lower, depending on the group of cyanotoxins targeted (MacKenzie and others, 2004; Guo and others, 2017; Roy-Lachapelle and others, 2021). In the FLX study (Johnston and others, 2025), microcystin concentrations were consistently lower when measured using MS compared to ELISA (Box 3—Analysis of Solid Phase Adsorption Toxin Tracking Extracts by Enzyme-linked Immunosorbent Assay and Mass Spectrometry Methods; fig. 8). Anatoxin, however, showed consistently higher concentrations measured by MS compared to ELISA (Box 3—Analysis of Solid Phase Adsorption Toxin Tracking Extracts by Enzyme-linked Immunosorbent Assay and Mass Spectrometry Methods; fig. 8). These patterns also can be seasonally dependent (Birbeck and others, 2019). Agreement between ELISA and MS results occurred through summer months, but by fall, values diverged, with MS values remaining lower than the ELISA results (Birbeck and others, 2019). Birbeck and others (2019) hypothesized that biodegradation products accumulated in the fall after the summer bloom period, resulting in overestimated cyanotoxin concentrations from the ELISA methods.

As with ELISA, matrix interference issues can occur when analyzing SPATT extracts using MS methods. Analysts working with extracts from the SSD study (table 2) determined that samples extracted in methanol acidified with formic acid failed quality assurance/quality control (QA/QC) checks (Box 4—Reducing Matrix Effects to Improve Cyanotoxin Detection). Analysts had to troubleshoot and alter their protocols to limit the time microcystin molecules were exposed to formic acid through the extraction process (Box 4—Reducing Matrix Effects to Improve Cyanotoxin Detection). Although analysts may be able to anticipate matrix interference issues and make needed adjustments to the analytical process, there may be unknown compounds in the sample matrix that could interfere with the analysis and bias the result. For example, if electrospray ionization is used in the mass spectrometer, it is possible that salts in the sample may affect ionization efficiency because they are nonvolatile and likely not detected in the same concentration as in the calibration standards (King and others, 2000), which can result in ionization suppression or possibly ionization enhancement, biasing the results (George and others, 2018). Although matrix interference issues are part of most analytical methods, the relatively recent (2004) advent of SPATT technology means that there is limited knowledge of possible interferences and thus greater uncertainty in analytical results. Like ELISA methods and the SSD study, it is likely that future laboratory experimentation and method development will reduce matrix interference issues.

Relating Solid Phase Adsorption Toxin Tracking Data to Discrete Cyanotoxin and Ancillary Data

Directly relating cyanotoxin data collected using SPATT samplers to other discrete cyanotoxin, environmental, genetic, or phytoplankton community datasets collected before, during, or after the deployment period is a limitation. Efforts have been made to determine adsorption and desorption kinetics of several cyanotoxin and resin combinations in laboratory settings, but performance is not consistent once samplers are deployed into variable field environments (MacKenzie and others, 2004; Lane and others, 2010; Zhao and others, 2013; Kudela, 2017; Roué and others, 2018). Without accurate adsorption/desorption rates, calibrating SPATT samplers to units used in discrete sampling or regulatory contexts, like the mass of cyanotoxin per volume of medium (for example, micrograms of cyanotoxin per liter of water [µg/L]), is not possible. Similarly, when benthic periphyton or other tissues are analyzed, cyanotoxin concentration is reported as mass cyanotoxin per mass of dry weight or ash-free dry weight of the medium (for example, µg/g dry weight) and cannot be directly compared to SPATT results either. Concentrations from SPATT samplers generally are reported as the mass of cyanotoxin per gram of resin (ng/g) or adsorption rate during the total deployment period (ng/g per day). These differences in reporting units and unknown adsorption and desorption kinetics make it challenging to directly compare SPATT results to water or benthic sample concentrations (Lane and others, 2010; Kudela, 2011, 2017; Jia and others, 2022). However, because of the passive nature of SPATT samplers, they are often part of larger research or monitoring projects, and in those instances, results need to be indirectly compared and interpreted alongside other parameters.

Of the 12 USGS SPATT studies featured in this report (tables 2, 3), most include collection of discrete cyanotoxin samples when SPATT samplers are deployed or retrieved. Collecting discrete water samples at these times provides information about the conditions at the start or end of a SPATT sampler deployment period but could miss pulses at times in between. Discrete samples also could be collected within a SPATT sampler deployment period, which would provide information about how cyanotoxin concentrations varied while the sampler was deployed. Another reason SPATT samplers generally complement rather than replace discrete sampling is because SPATT samplers only adsorb dissolved cyanotoxins in the environment (Lane and others, 2010; Kudela, 2011, 2017; Roué and others, 2018). Tatters and others (2021) suggest that differences detected between discrete and SPATT sample concentrations could be the result of the cyanobacterial cells remaining intact, not allowing cyanotoxins to be released in the dissolved form into the environment. During the study, they detected cylindrospermopsins only in discrete samples and saxitoxins and nodularins only in the SPATT samples (Tatters and others, 2021). With variation in the fraction and concentration in an environment, using multiple lines of evidence from SPATT samplers and discrete cyanotoxin samples (especially if analyzing for total and dissolved cyanotoxins) can help tease apart the complexity of cyanotoxin dynamics in natural water-body systems. In the FLX study (table 2; Box 2—Solid Phase Adsorption Toxin Tracking Use to Determine Cyanotoxin Occurrence Across Multiple Depths; Box 4—Reducing Matrix Effects to Improve Cyanotoxin Detection), early season deployments provided some evidence of early cyanotoxin detection as the SPATT samplers integrated low levels of cyanotoxins earlier than they were detected using discrete sampling. Discrete samples may be used to help validate the relative cyanotoxin concentrations but because of the dynamic nature of cyanotoxin production, persistence, and transport, cyanotoxin content is highly variable, and the relation between passive samplers and discrete samples will not be static.

In addition to discrete cyanotoxin data, phytoplankton identification and enumeration, environmental DNA, including cyanotoxin synthetase gene quantification and metagenomics, and additional co-occurring contaminants or water-quality parameters can supplement and provide context to SPATT studies (Graham and others, 2012, 2020; Francy and others, 2015, 2016; Zuellig and others, 2021; Linz and others, 2023). SPATT data may be augmented with these additional types of samples to identify potential cyanotoxin producers, which may include planktonic or benthic cyanobacteria such as Microcystis, Oscillatoria, Microcoleus, Nostoc, Dolichospermum, and many others. These taxa may be concentrated, cultured, and tested for the production of cyanotoxins, which may help explain the occurrences detected using SPATT samplers. It is important to remember that like discrete cyanotoxin data, these supplementary methodologies also are discrete samples and thus representations of an ecosystem at one moment in time, so results must be interpreted carefully (Lane and others, 2010; MacKenzie, 2010; Rodríguez and others, 2011; Zendong and others, 2016; Roué and others, 2018). In addition, the relation among cyanotoxin occurrence, abundance of potential cyanobacterial producers, and quantity of cyanotoxin synthetase genes can be complex. Because of the documented unreliability between cyanobacteria presence and cyanotoxin production (Lane and others, 2010; MacKenzie, 2010; Rodríguez and others, 2011; Peacock and others, 2018; Roué and others, 2018), they are not considered substitutes for cyanotoxin sampling. Similarly, the presence of cyanotoxin synthetase genes in the environment does not guarantee that the genes are actively being expressed, and therefore producing cyanotoxins (Pacheco and others, 2016). Cyanotoxins and cyanotoxin synthetase genes may persist longer in the environment than the causative cyanobacteria, which can create misleading co-occurrences with other potential cyanotoxin-producing species (Graham and others, 2012, 2020; Otten and others, 2015; Pacheco and others, 2016; Preece and others, 2017). Despite the challenges in comparing discrete data with SPATT results, cyanotoxin synthetase gene analysis may provide useful compliments to cyanotoxin data (Graham and others, 2012, 2020; Zuellig and others, 2021).

Ancillary water-quality measurements like temperature, salinity, nutrient concentrations, or pH can provide context if questions regarding drivers of cyanotoxin concentration and production are of interest (Paerl and Otten, 2013; Francy and others, 2015; Matson and others, 2020; Tatters and others, 2021; Howard and others, 2022; Anderson and others, 2023). However, like discrete cyanotoxin data, phytoplankton identification and enumeration, and cyanotoxin synthetase gene quantification, these discrete data also suffer from the same snapshot-in-time challenge. To help combat this issue, sondes were deployed during the FLX study at the same depths as SPATT samplers (1 m, 13 m, and 26 m) to ensure continuous water-quality data were collected during each deployment. Supplementary discrete sampling could be done during a SPATT deployment but would require additional staff time and analysis costs. Measurements of photosynthetically active radiation or light intensity also can provide insight on photodegradation potential if SPATT samplers are deployed in direct sunlight near the surface of a water body (Kurtz and others, 2021). Together, discrete cyanotoxin concentrations, information from phytoplankton communities, as well as an understanding of the biogeochemical factors affecting a system, can provide context needed to help interpret cyanotoxin concentrations collected using SPATT samplers.

Quality Assurance and Quality Control

Quality-assurance and quality-control measures are an important part of laboratory best practices. As with many environmental studies, replicates, blanks, and spikes can be used to ensure data validity and help with SPATT data interpretation (U.S. Geological Survey, variously dated). The number of potential QA/QC samples will depend on the study objectives and finding a balance among cost, necessity, and robustness. Across the 12 USGS studies discussed in this report (tables 2, 3), QA/QC procedures were not consistent. Generally, studies that examine cyanotoxins using commercially available ELISA test kits follow published methods for SPATT sampler assembly and extraction (Negrey and others, 2023) and U.S. Environmental Protection Agency (EPA) Method 546 for the analysis of microcystins and nodularins (U.S. Environmental Protection Agency, 2016). Some studies (SRB and NAR) also have adopted the quality control requirements of EPA Method 546 for the analysis of other cyanotoxins (cylindrospermopsin, anatoxin-a, saxitoxin) by ELISA. Built into most commercially available ELISA test kits are several standards, a control, and a blank; all standards, controls, blanks, and samples are analyzed in duplicate or triplicate, and if any of the method-specified QA/QC measures are outside of the prescribed limits, results are not valid and must be reanalyzed. Concentrations below the lowest non-zero standard are reported as non-detects, although for some studies (SRB and NAR), concentrations below the second lowest non-zero standard were reported as non-detects, with concentrations between the lowest and second lowest non-zero standards noted as being between the detection limit and reporting limit. In the NJC study (table 2), samples were analyzed using these ELISA methods, but a subset of extracts (between 5 and 20 percent) were sent to an outside laboratory for additional MS analysis to ensure validity of the data. As described in Box 4—Reducing Matrix Effects to Improve Cyanotoxin Detection, the FLX study (table 2) scientists used a combination of ELISA and MS analyses to compare results obtained from three analytical methods.

In addition to analytical controls (Box 3—Analysis of Solid Phase Adsorption Toxin Tracking Extracts by Enzyme-linked Immunosorbent Assay and Mass Spectrometry Methods), field replicates are an important component of robust QA/QC programs. Field replicates typically are deployed in the same location and for the same amount of time as field samplers to test for variability in the samplers themselves. As a best practice, SPATT samplers are deployed during most USGS studies in duplicate. In instances where both SPATT samplers are successfully recovered, the additional SPATT sampler may be analyzed as a field replicate. During the SRB study, replicate SPATT hoops were analyzed for microcystins, cylindrospermopsins, and saxitoxins and mean relative percentage differences of 13.5, 16.3 and 8.8 percent were detected, respectively, from six or seven sampling sites (Spitz and others, 2025). In other instances, increased variation may be expected between SPATT field replicates if there are variations in sampler construction (for example, sampler style, size, amount of resin used, and so on) or subtle variations in the flow around the multiple samplers deployed in each environment.

Blanks can be used to account for analytical interference from any contamination that may occur during sampler construction, deployment, processing, or analysis. Fabrication blanks are built concurrently with the field-deployed SPATT samplers but remain frozen in the laboratory until extraction. Extraction would then happen at the same time and in the same manner as the field samplers to account for any contamination during the construction, storage, extraction, or analysis steps. Separately, field blanks are built concurrently with field-deployed SPATT samplers and transported to the field sites in the same manner as the field samplers but are not deployed. In most cases, SPATT samplers are transported in bags filled with organic free water on ice. Upon arrival at the field site, field blanks are removed and exposed to the air for about the same amount of time it takes to place the field SPATT sampler into the water. This field blank can account for any potential contamination during transportation or deployment. Field blanks also can account for any potential deposition of aerosolized cyanotoxins during deployment or retrieval. And finally, laboratory blanks are extracted at the same time as SPATT samplers to account for any contamination during the extraction, evaporation, or reconstitution steps. During the SRB and FLX studies, laboratory blanks were processed the same way as the field samples, but the extraction column did not contain any resin beads. There was no contamination in extraction blanks using this QA/QC process during the FLX study. Although a rare occurrence, of the 17 laboratory blanks analyzed for the SRB study, 1 cylindrospermopsin blank failed and was above the 0.05 µg/L QA/QC threshold, indicating possible contamination during the extraction, evaporation, or reconstitution steps. In contrast, the ODW study included one laboratory blank per extraction batch and never had a blank with a positive detection.

Lastly, spikes can be used to determine recovery of the targeted analyte and examine potential analytical interference and effects on identification and quantitation. Matrix spikes are prepared with a known concentration of the targeted cyanotoxin, which then go through all the same processing and analysis steps as an environmental sample. Matrix spikes can help determine if the sample matrix affects detection of the targeted cyanotoxin. Separate matrix spikes must be prepared to account for each cyanotoxin of interest. Additional procedural spikes can determine the recovery of a targeted cyanotoxin during specific processing step, or throughout the process. This type of spike would be made without the SPATT sample matrix. After discussions with the laboratory doing the analyses, procedural spikes were used to evaluate samples from the SSD study (table 2). After every 20 samples were analyzed, an ultrapure water sample spiked to a known concentration (1 part per billion) was examined to ensure the instrument stayed within QA/QC thresholds. Ensuring a combination of replicates, blanks, and spikes are collected and analyzed during a study can ensure accurate and valid SPATT data are collected.

Information Gaps and Key Research Considerations

In spite of the progress that has been made since SPATT samplers were developed in 2004 (MacKenzie and others, 2004), there are many information gaps that hamper their widespread adoption into monitoring programs. There is little guidance that can provide consistent use and data quality. Across the USGS, SPATT protocols (table 3) are primarily based on procedures published by Negrey and others (2023) and information included in reviews from Roué and others (2018) and Kamali and others (2022). The recommendations in these sources vary in terms of SPATT preparation, deployment, and extraction guidance, and therefore further methodological research is warranted to ensure SPATT data are being produced and interpreted consistently. Through preparation of this report, USGS researchers identified seven priority research and monitoring themes, each with specific questions that when answered, may improve the use of SPATT samplers in the future (table 4). For example, research providing information about how deployment and storage time (before extraction) affects SPATT sampler results (table 4) would be helpful in the interpretation of SPATT data. Deployment periods in the literature range from hours (Wood and others, 2011), to weeks (Peacock and others, 2018), to monthly intervals (Anderson and others, 2023). Similar variation in deployment length occurred within the USGS studies discussed throughout this report, with deployment periods that ranged from minutes to monthly intervals (table 3). It is unlikely that there is a single, optimal deployment period for SPATT samplers, given varying aquatic environments and study objectives. However, determining the factors (for example, previously collected discrete cyanotoxin data or flow rates, as suggested by the SRB study; Box 1—Assessing the Impact of Deployment Length on Solid Phase Adsorption Toxin Tracking Concentrations) that influence deployment time would inform future study designs.

Questions remain about cyanotoxin adsorption and desorption kinetics to and from the SPATT sampler resin in relation to deployment duration and the environmental conditions. Research has focused on the desorption kinetics of cyanotoxins off of the SPATT resin using a variety of solvents for analysis by MS or ELISA (Fux and others, 2008; Kudela, 2011; Wood and others, 2011; Zhao and others, 2013); however, natural desorption of cyanotoxins in a variety of environments and hydrologic conditions has yet to be fully explored. Questions also remain regarding the effect of long-term storage on SPATT samplers frozen after retrieval and extraction (Kudela, 2017). Wood and others (2011) did not detect anatoxin degradation products after extraction of samples frozen at −20 °C, suggesting their stability. Negrey and others (2023) examined the consequences of storing extracts loaded with anatoxins, microcystins, and nodularins stored at −20 °C and −80 °C for as much as 90 days and determined no statistical difference in percentage recovery of all cyanotoxins. However, SPATT samplers often are stored for periods longer than 90 days, and understanding longer term storage effects (for example, as much as 6 months) on additional cyanotoxins of interest (for example, cylindrospermopsins, saxitoxins) will further improve the quality of SPATT data.

Biofouling and photodegradation of adsorbed cyanotoxins also require further examination. The literature suggests that biofouling can affect passive samplers (Richardson and others, 2002; Lane and others, 2010), but to our knowledge, this has not been examined in detail using SPATT samplers. Fouling has the potential to bias SPATT sampler results up or down, therefore understanding the effects of fouling warrants further study. Similarly, direct and indirect photodegradation, phenomena known to degrade cyanotoxins at the water surface (Parker and others, 2016; Kurtz and others, 2021), could affect cyanotoxins once they are sorbed onto SPATT resin in the environment. A better understanding of photodegradation effects to SPATT derived cyanotoxin concentrations could allow for improvements in deployment strategies (for example, ensuring SPATT samplers are shaded if deployed at the surface) and increased measurement accuracy. Other priority information gaps identified include SPATT sampler design, resin activation periods, efficient extraction methods, and QA/QC (table 4). Addressing these knowledge gaps through field and laboratory experimentation is likely to provide new information, refine SPATT methods, and ensure consistent generation and interpretation of SPATT data across USGS and the broader cyanotoxin research community.

Integrating Solid Phase Adsorption Toxin Tracking into Research and Monitoring Programs

Integrating SPATT sampling techniques into established and developing cyanotoxin research and monitoring programs will allow for improved understanding and management of freshwater, brackish, and marine ecosystems (Kudela, 2017; Howard and others, 2022). Three key components of an effective HAB monitoring strategy described by Howard and others (2022) include ensuring efforts are coordinated, detecting multiple toxin groups, and using a combination of discrete and passive sampling techniques to fully capture toxin dynamics. An excellent example of SPATT implementation into a coordinated monitoring program is the nationwide NPS study (table 2; Box 5—Implementation of Solid Phase Adsorption Toxin Tracking Monitoring into Nationwide Harmful Algal Bloom Monitoring Toolbox). This study, part of the USGS-NPS Water Quality Partnership, was established with the goal of improving the toolbox available to park staff to identify and respond to algal blooms and toxin events in national park units across the United States (NPS study; Box 5—Implementation of Solid Phase Adsorption Toxin Tracking Monitoring into Nationwide Harmful Algal Bloom Monitoring Toolbox). U.S. Geological Survey researchers provided guidance to enable park staff and volunteers to deploy and retrieve SPATT samplers. The use of SPATT samplers allowed park staff to look for multiple cyanotoxins, pair discrete toxin and environmental data with their SPATT deployments and use consistent methodology across the 22 national park units (2025; Box 5—Implementation of Solid Phase Adsorption Toxin Tracking Monitoring into Nationwide Harmful Algal Bloom Monitoring Toolbox; fig. 1).

Inclusion of SPATT samplers in research programs can be an effective way to better understand cyanotoxin occurrence, drivers, transport, and composition in lentic, lotic, and benthic ecosystems. Keeping in mind that SPATT and discrete cyanotoxin sampling techniques are complimentary and not interchangeable, this combination of techniques can provide multiple lines of evidence to better understand cyanotoxins in dynamic systems. In lotic systems like rivers or estuaries where water movement from tides, winds, or river flows can transport cyanotoxins long distances, time-integrating SPATT samplers can capture downstream cyanotoxin events that could otherwise be missed (Kudela, 2017). For example, deploying SPATT samplers up and downstream could provide insights for potential transport or degradation that may occur in a lotic system. Bouma-Gregson and others (2018) detected consistently higher anatoxin and microcystin accumulation upstream, which, when analyzed with other data collected during this study, indicated that increased discharge or reduced cyanotoxin production likely resulted in the lower accumulation downstream, depending on the site. Similarly, in the Southern California Bight, SPATT samplers were deployed along several creeks and rivers flowing into the estuary to examine cyanotoxin dispersal and sources (Tatters and others, 2019). Using a combination of discrete and SPATT sampling techniques, Tatters and others (2019) detected elevated toxin concentrations downstream, which indicated that transport may play an important role in cyanotoxin dispersal in this system. The use of SPATT samplers in this study also helped demonstrate the need for a thorough and coordinated cyanotoxin monitoring program in this estuary. These studies (Bouma-Gregson and others, 2018; Tatters and others, 2019) also highlight the importance of considering benthic sources of cyanotoxins to the water column. The ODW study also detected benthic cyanobacteria in surface net tow samples associated with SPATT deployments. The genera Nostoc and Tolypothrix were detected in a large number of samples that also contained neurotoxins (Carpenter and Wise, 2023; Carpenter and others, 2025), further indicating that mixing or migration of benthic cyanobacteria to the water column may contribute to detections reported using SPATT samplers (McCarthy and others, 2014).

In lentic systems, SPATT samplers can be used to identify cyanotoxin dynamics at a fixed location. This identification can help to understand longitudinal patterns in cyanotoxin concentrations and, for lakes and reservoirs, the occurrence of cyanotoxins at various depths. These data can be one of multiple lines of evidence that shed light on the patterns of occurrence and drivers within a water body or watershed. Wiltsie and others (2018) deployed SPATT samplers in a lake used as a source of drinking water to examine drivers and seasonal patterns of cyanotoxin occurrence and composition. Multiple cyanotoxins were detected at both SPATT sampler locations where SPATTs were deployed year-round, with more frequency compared to accompanying discrete samples. The FLX study also demonstrated the effective use of SPATT sampling to examine seasonal cyanotoxin concentrations through the water column (Box 4—Reducing Matrix Effects to Improve Cyanotoxin Detection). This information is especially important to consider in deep lakes or reservoirs where seasonal stratification occurs. As demonstrated throughout this report, using examples from 12 USGS projects (table 2), SPATT samplers can complement research programs in numerous ways. With continued laboratory and field research to fill known knowledge gaps and improve the use of SPATT samplers (table 4), incorporation into monitoring programs and other investigations can become more consistent and widespread.