Passive Sampling Methods Research Articles

Estimating the uncertainty in the atmospheric ammonia concentration in an urban area by Ogawa passive samplers

Abstract Ammonia, one of the ambient gasses that require environmental monitoring, is typically measured using a passive sampling method. The present work presents an evaluation of the uncertainty according to the Guide to the Expression of Uncertainty in Measurement for the measurement of the atmospheric ammonia concentration as determined by an Ogawa passive sampler, using a colorimetric method. The analytical results report the uncertainty only as a standard deviation of repeated measurements, but not all sources of uncertainty are considered. In this work, the major sources of uncertainty in the measurements are identified as contributions to the linear least-square regression lines, repeatability and recovery. The result, including the expanded uncertainty (k = 2) at a level of confidence of 95%, is 39.2%. The aforementioned results indicate that the Ogawa sampler can be successfully deployed to estimate the atmospheric NH3 and could find wide application in environmental monitoring. However, to obtain correct conclusions, the uncertainty in the measurements must be considered.

Evaluating bee (Hymenoptera: Apoidea) diversity using Malaise traps in coffee landscapes of Costa Rica

AbstractEven though Arabica coffee (Coffea arabica Linnaeus, Rubiaceae) can self-pollinate, bees are important pollinators, without which there is lower fruit quality and yield. We studied bee diversity in coffee agroecosystems in Costa Rica during two coffee flowering seasons (2005 and 2006). Malaise traps were used as a passive sampling method to collect bees during coffee blooms. We collected 1012 bee individuals from three different site types: nonagricultural fields and shaded and unshaded coffee farms. Unshaded coffee farms had significantly higher species richness (S) and number of bee individuals (n) than did the shaded coffee farms and nonagricultural sites. Overall bee diversity did not differ among site types but evenness (J′) was significantly lower in unshaded coffee farms. Using a more detailed community analysis, there was a significant association between functional groups and habitat type with more species and individuals of small-bodied ground-nesting bees (Lasioglossum (Dialictus) Robertson) associated with unshaded coffee farms. A large proportion (49%) of bees collected were of this subgenus, which was never before reported as common in coffee agroecosystems. Further studies should establish whether Dialictus is important in coffee pollination. We propose strategies involving conservation of native bees through simple habitat management for small-scale coffee farms that may improve crop quality and quantity.

Passive Sampling of Polycyclic Aromatic Hydrocarbons in Johor Strait by Sol–gel Polydimethylsiloxane

Passive sampling technique is an alternative technique in the monitoring of organic pollutants in aqueous environment. This technique is able to overcome several drawbacks of the conventional grab sampling technique. In this study, passive sampler consists of glass tubes coated with sol–gel polydimethylsiloxane (PDMS) were fabricated in the laboratory. The passive samplers were used to study the level and distribution of 15 polycyclic aromatic hydrocarbons (PAHs) in Johor Strait. Performance of the passive sampler was tested in the laboratory with an exposure system that has similar flow rate, temperature and water salinity with Johor Strait. Then, the passive sampler was deployed to five sampling stations located in Johor Strait for 7 and 30 days to accumulate PAHs in seawater. High performance liquid chromatography with fluorescence detector was used to analyze PAHs in the samples. All PAHs spiked in the exposure system were found to accumulate in the PDMS coating. Mass of 15 PAHs extracted from the passive sampler after deployment to Johor Strait was in the range of 0.372–1.131 ng/PDMS glass tube. Findings obtained from passive sampling method and seawater analysis were comparable. Passive sampling technique is low cost, simple and less solvent consumption.

Using Polyacrylate-Coated SPME Fibers To Quantify Sorption of Polar and Ionic Organic Contaminants to Dissolved Organic Carbon

A passive sampling method using polyacrylate-coated solid-phase microextraction (SPME) fibers was applied to determine sorption of polar and ionic organic contaminants to dissolved organic carbon (DOC). The tested contaminants included pharmaceuticals, industrial chemicals, hormones, and pesticides and represented neutral, anionic, and cationic structures. Prior to the passive sampler application, sorption of the chemicals to the fibers was characterized. This was needed in order to accurately translate concentrations measured in fibers to freely dissolved aqueous concentrations during the sorption tests with DOC. Sorption isotherms of neutral compounds to the fiber were linear, whereas isotherms of basic chemicals covered a nonlinear and a linear range. Sorption of acidic and basic compounds to the fiber was pH-dependent and was dominated by sorption of the neutral sorbate species. Fiber- and DOC-water partition coefficients of neutral compounds were both linearly related to octanol-water partition coefficients (log Kow). The results of this study show that polyacrylate fibers can be used to quantify sorption to DOC of neutral and ionic contaminants, having multiple functional groups and spanning a wide hydrophobicity range (log Kow = 2.5-7.5).

Determination of oil–water partition coefficients of polar compounds: silicone membrane equilibrator vs. SPME passive sampler

Experimental determination of oil-water partition coefficients often poses difficulties associated with emulsion formation. The aim of this work was to find an appropriate technique for determination of oil-water partition coefficients of polar, nonvolatile compounds. Two different methods were tested. The first method used a "silicone membrane equilibrator." For the second method, solid-phase microextraction (SPME) fibers with a polyacrylate (PA) coating were used as a passive sampler. With both methods, oil-water partition coefficients for 14 compounds with polar functional groups were determined at 37 °C with good repeatability (standard deviation 0.11 log units or lower). The partition coefficients determined with the silicone membrane equilibrator method ranged from 0.50 to 3.49 log units. The oil-water partition coefficients obtained with the PA-SPME passive sampling approach were significantly higher than those obtained with the silicone membrane equilibrator method for nine of 14 compounds. The differences were up to 0.39 log units (i.e., a factor of 2.5). Additional experiments suggested that this difference occurred because the sorption properties of the PA fibers used were influenced by the surrounding phase, e.g., through swelling of the polymer phase. Therefore, the SPME passive sampling method using PA fibers seems to be less reliable, whereas the silicone membrane equilibrator method was found to be a convenient technique for the determination of oil-water partitioning.

How Redox Conditions and Irradiation Affect Sorption of PAHs by Dispersed Fullerenes (nC60)

Surface properties, dispersion state, and sorption behavior of carbon-based nanomaterials will change after being released into the environment. To study these processes, five different scenarios were considered to probe the impact of changes in surface properties of dispersed fullerenes (nC60) on their sorption potential due to irradiation and presence of oxygen. Sorption isotherms of pyrene by nC60 were determined at environmentally relevant concentrations applying a passive sampling method. Isotherms of all dispersion scenarios were best fit with the Dubinin-Ashthakov model. Sorption was strongest for nC60 kept under anoxic condition. Both the presence of oxygen and irradiation significantly decreased the sorption capacity of nC60, while commercially available polyhydroxy fullerenes had the smallest sorption. In addition, competition for sorption sites was never observed in multiple sorbate experiments with four polycyclic aromatic hydrocarbons at small concentration. A strong relationship between sorption coefficients and hydrophobic properties of sorbates suggests that hydrophobic interactions are of major importance. The results emphasize that aging of released fullerenes results in a reduced strength of interactions with nonpolar compounds and, thus, reduces the impact on the environmental transport of hydrophobic pollutants.

Determination of BTEX concentration in the groundwater of east region of Isfahan using passive sampling method

Aims: This study was conducted to determine Benzene, Toluene, Ethylbenzene, and Xylene (BTEX) concentration regarding broken petroleum transmission pipeline in the east region of Isfahan, using passive sampling method. Materials and Methods: Dialysis bags as passive sampling devices with molecular cut-off diameter of 12000 Dalton, 3.2 cm width and around 12 cm lengths were used that contain 2 ml of olive oil as an absorbent. These devices were placed, in 10 sampling wells for 45 days. In order to determine the absorption efficiency of these devices, laboratory calibration operations were performed at the same time. Nitrate as an indicator of contamination of groundwater with chemical fertilizers and also electrical conductivity (EC) as an indicator for cations and anions concentration was measured. Results: The laboratory absorption efficiencies of BTEX were found to be 90%, 85%, 79%, and 81%, respectively. Concentration of these compounds in field samples were less than 10 μg/l. Nitrate concentration varied between 2.1 to 69.2 mg/l, and EC varied between 1.7 to 15.1 mS/cm.Conclusion: BTEX compounds were absorbed with high efficiency in the laboratory test, but their concentrations in wells were found to be negligible. Average concentration of nitrate in the wells was less than water quality guidelines, and had no relation with the direction of movement or depth of water. EC tended to increase in the direction of groundwater movement, but there were no correlations among depth of water, nitrate concentration, and mean of EC.

Temporal variation and spatial distribution of PAH in water of Three Gorges Reservoir during the complete impoundment period

Bioavailable concentrations of polycyclic aromatic hydrocarbons (PAHs) were investigated in water of Three Gorges Reservoir (TGR) using semipermeable membrane devices during the period of completely impounding water. ∑PAH concentrations in water of TGR in the period of completely impounding water were 15-381 ng L(-1). ∑PAH concentrations increased from town or counties to big industrialized cities in TGR, indicating urbanization effects on PAH pollution in the water. Tributaries in TGR have a certain contribution of PAH pollution to the mainstream of Yangtze River and their pollution could not be neglected. An obvious decrease of PAH concentration was observed after 175-m water impounding in 2011 in TGR. Several factors may account for this decrease, including execution of comprehensive treatment and management measures in TGR, less rainfall in 2011, and sedimentation effect caused by the dam. Passive sampling method has been successfully applied in the investigation of trace PAH in water of TGR and proved to be a useful and efficient tool for the management and sustainable development of the big reservoir. The results of the study provide valuable information about PAH pollution in the whole reservoir including some tributaries, and the pollution status is dynamically related with human activities. Therefore, PAH could be used as a marker compound or indicator in the network monitoring system to surveil and trace the pollution status in TGR.

Overview of passive Chemcatcher sampling with SPE pretreatment suitable for the analysis of NPEOs and NPs

The European Union Water Framework Directive (WFD; 2000/60/EC) is an important piece of environmental legislation that protects rivers, lakes, coastal waters and groundwaters (EC 2000). The implementation of the WFD requires the establishment and use of novel and low-cost monitoring programmes, and several methods, e.g. passive sampling, have been developed to make the sampling process more representative compared to spot sampling. This review considers passive sampling methods focusing mainly on a passive sampler named Chemcatcher®, which has been used for monitoring several harmful compounds in aquatic environments. Also, the sample treatment and analysis of nonylphenol ethoxylates (NPEOs) and nonylphenol (NPs) from water using solid phase extraction (SPE) is briefly summarized. The procedure of Chemcatcher passive sampling is quite similar to that of the SPE extraction since it concentrates the studied compounds from water as well. After sampling, the accumulated substances are extracted from the receiving phase of the sampler. The concentrations of NPEOs and NPs are currently monitored by taking conventional spot samples; SPE can be successfully used as a pretreatment procedure. Chemcatcher® passive sampling technique is a simple and useful monitoring tool and can be applied to new chemicals, such as NPEOs and NPs in aquatic environments.

A physicochemical model of sorption processes in NO2 passive sampling with air humidity effects

Aqueous triethanolamine (TEA) solutions are widely used as sorption medium for passive sampling of ambient NO(2), with NO(2) trapped and accumulated as nitrite ion. The results of test measurements of ambient NO(2) concentrations using passive sampling method showed that the simple approach commonly used to describe passive sampling process might lead to substantial systematic errors. Presented in the article is a new physicochemical model of the process of passive sampling of gaseous NO(2), with aqueous TEA solution used as a trapping medium. The model is based on the available results of experimental studies of interaction of gaseous NO(2) with TEA/water solutions. The key principles underlying the model are: (1) when absorbed by a trapping solution, NO(2) forms nitrite ion only on the condition that TEA is hydrated; (2) coefficient of conversion of NO(2) to NO(2)(-) is equal to one when reacting with hydrated TEA; and (3) the fraction of hydrated TEA molecules depends on air humidity at the moment of measurement. Validation of the model was made using the data of the field measurements carried out in the Middle Urals in 2007-2009. The new model was used to calculate average NO(2) concentrations. Concentrations calculated agreed well with the results obtained by reference methods. The difference between the datasets was statistically insignificant.

Air sampling procedures to evaluate microbial contamination: a comparison between active and passive methods in operating theatres

BackgroundSince air can play a central role as a reservoir for microorganisms, in controlled environments such as operating theatres regular microbial monitoring is useful to measure air quality and identify critical situations. The aim of this study is to assess microbial contamination levels in operating theatres using both an active and a passive sampling method and then to assess if there is a correlation between the results of the two different sampling methods.MethodsThe study was performed in 32 turbulent air flow operating theatres of a University Hospital in Southern Italy. Active sampling was carried out using the Surface Air System and passive sampling with settle plates, in accordance with ISO 14698. The Total Viable Count (TVC) was evaluated at rest (in the morning before the beginning of surgical activity) and in operational (during surgery).ResultsThe mean TVC at rest was 12.4 CFU/m3 and 722.5 CFU/m2/h for active and passive samplings respectively. The mean in operational TVC was 93.8 CFU/m3 (SD = 52.69; range = 22-256) and 10496.5 CFU/m2/h (SD = 7460.5; range = 1415.5-25479.7) for active and passive samplings respectively. Statistical analysis confirmed that the two methods correlate in a comparable way with the quality of air.ConclusionIt is possible to conclude that both methods can be used for general monitoring of air contamination, such as routine surveillance programs. However, the choice must be made between one or the other to obtain specific information.

Spatial and temporal analysis of ground level ozone and nitrogen dioxide concentration across the twin cities of Pakistan

The analyses presented in this paper include the concentration levels of NO2 and O3 measured during 2 successive years in twin cities (Rawalpindi and Islamabad) of Pakistan from November 2009 to March 2011. NO2 was determined using the passive sampling method, while ozone was determined by Model 400E ozone analyzer. The average NO2 and O3 concentration in twin cities of Pakistan was found to be 44 ± 6 and 18.2 ± 1.24ppb, respectively. Results indicate that the concentration of NO2 and O3 show seasonal variations. Results also depict that NO2 and O3 concentration levels are high in areas of intense traffic flow and congestion. Rawalpindi has more elevated levels of NO2 and O3 as compared to the Islamabad due to the narrow roads, enclosing architecture of road network and congestion. Climatic variables also influenced the NO2 and O3 concentration, i.e., temperature is positively related with O3, while negatively related with NO2, relative humidity is directly related with NO2 and inversely related with O3, whereas rainfall show negative association with both NO2 and O3 concentration. Comparing the results with WHO standards reveals that NO2 concentration levels at all the sampling points are above the permissible limit, while ozone concentration is still lower than the WHO standards. Thus, there is a need to take appropriate steps to control these continuously increasing levels of NO2 and O3 before they become a serious hazard for the environment and people living in those areas.

Dispersion State and Humic Acids Concentration-Dependent Sorption of Pyrene to Carbon Nanotubes

Sonication and humic acids (HA) are known to disperse carbon nanotube (CNT) suspensions, but potential effects on sorption of chemicals to CNTs remain poorly understood. We applied a passive sampling method to investigate the influence of dispersion/aggregation on sorption of pyrene to CNTs. Sonication broke down CNT aggregates and increased pyrene sorption affinity by up to 1.39 orders of magnitude. Sorption surfaces newly exposed by sonication remained available to pyrene even after reaggregation occurred, suggesting an irreversible effect of sonication. The presence of HA decreased sorption of pyrene to CNTs, but at the highest HA concentration investigated (200 mg/L), sorption affinity was still 1.90 orders of magnitude larger than sorption of pyrene to HA alone. Specific interactions between pyrene and CNTs were thus still taking place, in spite of the presence of a HA coating on the CNTs' surface. A greater suppression of sorption by CNTs occurred when the HA addition was combined with a sonication pretreatment. Sorption isotherm fitting indicated that the maximum sorption capacity, sorption affinity, and heterogeneity of the CNT surface were all affected by sonication and the presence of HA at a concentration as low as 1 mg/L. The present results contribute to an improved understanding of the sorption behavior of CNTs in both natural and wastewater systems.

A Comparison of Terrestrial Arthropod Sampling Methods

Abstract: Terrestrial arthropods are extremely important ecosystem components. The choice of best approaches to collect the wide range of terrestrial arthropods has been a topic of long-lasting debates. This article provides a brief overview of common sampling methods for terrestrial arthropod assemblages. We divide sampling methods into three main categories: passive sampling methods without any “activity density” bias, passive sampling methods with an “activity density” bias, and active sampling methods with inherent “activity density” and often further species-dependent biases, discussing their individual advantages and shortcomings as basis for biodiversity studies and pest control management. The selection of the optimal sampling methods depends strongly on the purpose of individual studies and the ecology and behavior of the arthropod groups targeted. A combination of different suitable methods is highly recommended in many cases.

Microbial air monitoring in operating theatres: experience at the University Hospital of Parma

Microbial air monitoring in operating theatres has been a subject of interest and debate. No generally accepted sampling methods and threshold values are available. To assess microbial air contamination in empty and working conventionally ventilated operating theatres over a three-year period at the University Hospital of Parma, Italy. Air sampling was performed in 29 operating theatres. Both active and passive sampling methods were used to assess bacterial and fungal contamination. In empty theatres, median bacterial values of 12 colony-forming units (cfu)/m(3) [interquartile range (IQR) 4-32] and 1 index of microbial air contamination (IMA) (IQR 0-3) were recorded. In working theatres, these values increased significantly (P<0.001) to 80cfu/m(3) (IQR 42-176) and 7 IMA (IQR 4-13). Maximum recorded values were 166cfu/m(3) and 8 IMA for empty theatres, and 798cfu/m(3) and 42 IMA for working theatres. Combining active and passive samplings, fungi were isolated in 39.13% of samples collected in empty theatres and 56.95% of samples collected in working theatres. Over the three-year study period, bacterial contamination decreased in both empty and working theatres, and the percentage of samples devoid of fungi increased. In working theatres, a significant correlation was found between the bacterial contamination values assessed using passive and active sampling methods (P<0.001). Microbiological monitoring is a useful tool for assessment of the contamination of operating theatres in order to improve air quality.

In-vivo passive sampling to measure elimination kinetics in bioaccumulation tests

The application of in-tissue passive sampling to quantify chemical kinetics in fish bioconcentration experiments was investigated. A passive sampler consisting of an acupuncture needle covered with a PDMS tube was developed together with a method for its deployment in rainbow trout. The time to steady state for chemical uptake into the passive sampler was >1d, so it was employed as a kinetically limited sampler with a deployment time of 2h. The passive sampler was employed in parallel with the established whole tissue extraction method to study the elimination kinetics of 10 diverse chemicals in rainbow trout. 4-n-nonylphenol and 2,4,6-tri-tert-butylphenol were close to or below the limit of quantification in the sampler. For chlorpyrifos, musk xylene, hexachlorobenzene, 2,5-dichlorobiphenyl and p,p′-DDT, the elimination rate constants determined with the passive sampler method and the established method agreed within 18%. Poorer agreement (35%) was observed for 2,3,4-trichloroanisole and p-diisopropylbenzene because fewer data were obtained with the passive sampling method due to its lower sensitivity. The work shows that in-tissue passive sampling can be employed to measure contaminant elimination kinetics in fish. This opens up the possibility of studying contaminant kinetics in individual fish, thereby reducing the fish requirements and analytical costs for the determination of bioconcentration factors.

Comparison of different triolein-based composite semipermeable membranes for passive sampling of organochlorine pesticides

Passive sampling methods have been widely used for more than 20 years because they can provide information on time-weighted average concentrations (C TWA) of hydrophobic organic contaminants in water. Many factors affect the efficiency and applicability of a passive sampling process. Among others, selection of the sampler design and choice of the supporting membrane material are important factors that determine the performance of passive samplers. In this work, the influences of some critical properties of membranes, such as membrane density, hydrophilicity of the membrane surface, and chemical and mechanical resistances, on the uptake of organic chlorinated pesticides (OCPs) were evaluated. Four membrane materials, namely cellulose acetate (CA), cellulose acetate butyrate (CAB), a cellulose acetate-cellulose triacetate blend (CA-CTA), and cellulose triacetate (CTA), and their triolein-embedded samplers were selected for the study. The four membranes are all hydrophilic; CAB has the highest contact angle and CA has the lowest. CTA is chemically and mechanically resistant, and has better tensile strength than the other membranes have. CAB does not have good chemical resistance and dissolves in hexane/dichloromethane solvent. CA and CTA alone showed lower uptake rates for the selected OCPs than with triolein-embedded passive samplers, indicating that triolein increased the accumulation capacity. Triolein-embedded CA (TECAM), CA-CTA (TECA-CTA), CTA (TECTA), and CAB (TECAB) as passive samplers showed similar uptake patterns; TECTA showed the highest uptake rates and TECAB showed the poorest accumulation. In summary, TECTA gave the best performance among all the tested composite semipermeable membranes.

The measurement time required for determining total NH3 losses after field application of slurries by trail hoses

SUMMARYMonitoring ammonia (NH3) emission is time consuming and requires specialized measurement equipment. The measurement time can be reduced if there is a close relationship between time and subsequent cumulated NH3 emission values. A statistical analysis was employed to study the relationship between cumulative NH3 emissions over varying time intervals and final NH3 loss after 3 days of measurement. A large number of multi-plot field experiments on NH3 loss after the application of animal and biogas slurries by trail hoses to crops in Northern Germany were carried out from 2007 to 2010. Based on data from 2007, measured using a passive sampler method, a linear empirical model was developed to calculate final cumulated NH3 loss from intermediate cumulated losses. Linear model fitting showed that cumulative NH3 losses after 24 h were significantly correlated with final cumulated NH3 losses, explaining more than 0·98 of its variation. The linear coefficient was 1·34, implying that c. 0·73 of final NH3 loss occurred within the first 24 h. Validation by datasets obtained from another year (2008), two additional measurement methods and another agro-region (marsh area, 2009/10) resulted in a close agreement of model predictions with measured data within the range of model uncertainty and data variation. The results underpin the feasibility of calculating final NH3 losses from cumulative losses during first 24 h after slurry application and can be used to simplify NH3 loss measurement after the application of liquid slurries in multi-plot field experiments. The slope of the linear relationship is only valid for liquid slurries and the environmental conditions of the present study, which are typical for many agro-regions in north-western Europe, and will have to be adapted for different climatic conditions. A time-efficient measurement of emissions from solid organic fertilizers might require a different time span.

Predicting sample size required for classification performance

BackgroundSupervised learning methods need annotated data in order to generate efficient models. Annotated data, however, is a relatively scarce resource and can be expensive to obtain. For both passive and active learning methods, there is a need to estimate the size of the annotated sample required to reach a performance target.MethodsWe designed and implemented a method that fits an inverse power law model to points of a given learning curve created using a small annotated training set. Fitting is carried out using nonlinear weighted least squares optimization. The fitted model is then used to predict the classifier's performance and confidence interval for larger sample sizes. For evaluation, the nonlinear weighted curve fitting method was applied to a set of learning curves generated using clinical text and waveform classification tasks with active and passive sampling methods, and predictions were validated using standard goodness of fit measures. As control we used an un-weighted fitting method.ResultsA total of 568 models were fitted and the model predictions were compared with the observed performances. Depending on the data set and sampling method, it took between 80 to 560 annotated samples to achieve mean average and root mean squared error below 0.01. Results also show that our weighted fitting method outperformed the baseline un-weighted method (p < 0.05).ConclusionsThis paper describes a simple and effective sample size prediction algorithm that conducts weighted fitting of learning curves. The algorithm outperformed an un-weighted algorithm described in previous literature. It can help researchers determine annotation sample size for supervised machine learning.

DGT Measurement of Dissolved Aluminum Species in Waters: Comparing Chelex-100 and Titanium Dioxide-Based Adsorbents

Aluminum is acutely toxic, and elevated concentrations of dissolved Al can have detrimental effects on both terrestrial and aquatic ecosystems. Robust analytical methods that can determine environmentally relevant Al fractions accurately and efficiently are required by the environmental monitoring community. A simple, robust passive sampling method, the diffusive gradients in thin films (DGT) technique, was evaluated for the measurement of dissolved Al species in freshwater and marine water using either Chelex-100 or Metsorb (a titanium dioxide-based binding agent) as the adsorbent. Mass vs time DGT deployments at pH 5.05 (Al(3+) and Al(OH)(2+) dominate) and 8.35 (Al(OH)(4)(-) dominates) demonstrated linear uptake of Al (R(2) = 0.989 and 0.988, respectively) for Metsorb. Similar deployments of Chelex-DGT showed linear uptake at pH 5.05 (R(2) = 0.994); however, at pH 8.35 the mass of Al accumulated was 40-70% lower than predicted, suggesting that Chelex-100 is not suitable for Al measurements at high pH. The Metsorb-DGT measurement was independent of pH (5.0-8.5) and ionic strength (0.001-0.7 mol L(-1) NaNO(3)), whereas the Chelex-DGT measurement was only independent of ionic strength at pH 5.0. At pH 8.4, increasing ionic strength led to considerable underestimation (up to 67%) of Al concentration. Deployments of Metsorb-DGT (up to 4 days) in synthetic freshwater (pH range 5.4-8.1) and synthetic seawater (pH 8.15) resulted in linear mass uptakes, and the concentration measured by DGT agreed well with solution concentrations. Conversely, deployment of Chelex-DGT in synthetic seawater and freshwater (pH ≥7.7 Al(OH)(4)(-) dominant species) resulted in a decrease in accumulated mass with increasing deployment time. In situ field evaluations in fresh, estuarine, and marine waters confirmed that Metsorb-DGT was more accurate than Chelex-DGT for the measurement of dissolved Al in typical environmental waters.