Metal-series data guidance

This guidance is for data stewards uploading metal-series data to DataStream, to support a consistent approach when handling this data.

DataStream’s data schema (DS-WQX) aligns with the U.S. Environmental Protection Agency and U.S. Geological Survey developed Water Quality eXchange (WQX) data standard. This guidance is also informed by the Best Practices for Submitting Metals Data to the Water Quality eXchange (WQX).

For the purpose of this guidance, the term ‘metal-series’ applies to metals (e.g., Fe, Cr, Cu, etc.), metalloids (e.g., As, Si, etc.), as well as additional environmentally relevant trace elements that fall outside these categories (e.g., non-metals, lanthanides and actinides).

Table 1: Commonly analyzed, environmentally relevant metal-series analytes.

Data uploaded to DataStream needs to be formatted in the DataStream data structure (DS-WQX) which specifies required fields and allowed values for consistent vocabulary. Information on each characteristic (analyte) should be reported using the fields in Table 2.

Table 2: DataStream fields used to report a given metal-series characteristic.

The CharacteristicName identifies the characteristic (analyte) being measured. The full list of characteristic names is available in the DataStream Upload Template (CharacteristicName LOOKUP tab). For most metal-series analytes, the CharacteristicName is the analyte’s elemental name (e.g., “Aluminum” or “Lead”).

If you don’t see a CharacteristicName appropriate for the analyte in your dataset, let us know and we can add it to the list.

Redox-active species

For redox-active species with multiple stable oxidation states, the ionic form can be incorporated in the characteristic name (e.g. “Chromium (III)” and “Chromium (VI)”). If the oxidation state is unknown, the elemental name is acceptable (i.e., Chromium).

Isotopes

For metal-series analytes that exist as stable or radioactive isotopes, please refer to our Best Practices for Reporting Isotope Data.

Metals and ions

Some elements (e.g. Calcium, Magnesium, Sodium, and Potassium) can be measured either as ions, or in their elemental form in a metals-series. In both cases, the same CharacteristicName is applicable, and methods are differentiated in the ResultAnalyticalMethod field. For example, ion chromatography is a common method to measure ion concentrations, while mass spectrometry (e.g. ICP-MS) is a common method to measure elemental concentrations.

Phosphorus

The nutrient Phosphorus is a non-metal that is sometimes measured in its elemental form in a metal-series analysis suite. When this is the case, the preferred DataStream CharacteristicName is “Phosphorus, elemental”. Otherwise, in the majority of cases, the CharacteristicName for Phosphorus should be chosen based on the guidance in DataStream’s Best Practices for Reporting Nutrient Data.

Sample fractions describe the portion of the characteristic (analyte) that has been measured in a sample. An accurate representation of the sample fraction is important for interpretation and re-use of data.

Samples may be prepared using physical (e.g. filtration, decantation, and/or centrifugation) and/or chemical (e.g. extractions) techniques prior to an analysis protocol that quantifies or detects the analyte.

To select the correct sample fraction term, both the preparation and analysis method(s) should be considered.

The method(s) or protocol(s) reported by the laboratory may only include information about the analysis method, and not the preparation method. Because information about the analysis method only may not be sufficient to determine the sample fraction, we recommend consulting with the laboratory to identify which fraction term is appropriate in these cases.

To promote consistency and avoid ambiguity, DataStream requires the use of specific allowed values for sample fractions, based on the metal-series terminology recommended by WQX.

DataStream’s preferred sample fraction terms for water and sediment samples are explained below, and their definitions are provided in Tables 3 and 4.

For filtered water samples, the term ‘Dissolved’ refers to the fraction of the sample that passed through a filter (i.e., the filtrate). ‘Pot. Dissolved’ (i.e., potentially dissolved) is for samples that were filtered after treatment with nitric acid.

For unfiltered water samples, DataStream recommends using the terms ‘Extractable’ or ‘Total’ to indicate the fraction of the sample that has been analyzed following a chemical digestion or solubilization procedure. 'Total’ may be used if the procedure is expected to yield an analyte concentration that is representative of the whole sample, whereas the term ‘Extractable’ may be used in instances where weaker chemical extraction procedures were conducted.

The particulate portion of a water sample refers to the suspended particulate fraction that can be physically separated from the aqueous solution (e.g. by filtration, centrifugation, or decantation). For metal-series characteristics, we recommend using the sample fraction ‘Suspended’ for this particulate fraction of a water sample.

Sequential extractions, which may performed on the suspended particulate, will be discussed in a following section.

Sediment samples should be indicated in the ActivityMediaName either as “Surface Water Sediment” for suspended and/or shallow surface bed sediments or as “Subsurface Soil/Sediment” for deeper sediment (e.g. bed sediments or sediment cores). For more details, see the glossary tab of the DataStream Upload Template.

Metal-series characteristics (analytes) in sediments may be analyzed in their solid state, or they may be dissolved via an extraction process where they are exposed to reagents (e.g. salts or acids) that modify their solubility.

For sediment samples analyzed with a single (unique) extraction reagent, use the sample fractions ‘Total’ or ‘Extractable’. As with water samples, ‘Total’ is indicative of a strong (complete) extraction procedure, while ‘Extractable’ indicates a weaker (partial) extraction.

Sequential extractions, which may performed on sediment samples, will be discussed in the following section.

Table 3: Recommended sample fraction terms for submitting metal-series data to DataStream.

Metal-series characteristics may be analysed in a series of sequential extractions, where the sample is exposed to multiple reagents, and the analysis is repeated following each exposure. The measured value therefore indicates the amount of analyte associated with specific chemical fractions in the whole sample.

Sequential extraction analyses are most likely to be conducted on solid samples (e.g. sediments), or particulates in water samples.

Reagents may include salts (e.g. MgCl₂), organic acids (e.g. acetic acid), or inorganic acids (e.g. HCl). DataStream’s recommended terminology for describing sample fractions for sequential extractions (Table 4) is based on common operationally-defined terms from the literature (Tessier et al. 1979; Gleyzes et al. 2002).

For sediment samples, the sequential extraction process would be entered in the ResultSampleFraction column.

For the particulate portion of a water sample, the recommended ResultSampleFraction term is ‘Suspended’. As such, we recommend providing additional information about further sequential extraction procedures in the ResultComment column.

Table 4: Recommended terms for sequential extraction protocols.

* For sediment samples, these terms would be entered in the ResultSampleFraction column. For the particulate portion of water samples, these terms would be specified in the ResultComment column because the ResultSampleFraction should be reported as ‘Suspended’.

DataStream recommends using the ResultComment column to provide additional observation-level information, which can include further details regarding sample preparation methods and analyzed sample fractions.

For example, sample filtration most commonly occurs with a filter pore size of 0.45 µm. If the filter pore size was not 0.45 µm, this should be indicated in the ResultComment.

This field may also be used in instances where physical and chemical sample fractions cannot be accurately described using only the ResultSampleFraction field. For example:

• If ResultSampleFraction is ‘Suspended’ (i.e. the particulate portion of the water sample was analysed), ResultComment may be used to indicate additional chemical extraction procedures, if applicable (e.g., ‘Total’, ‘Extractable, oxide-bound’, etc.).
• If the ResultSampleFraction is ‘Extractable, other’, the ResultComment may provide further details on the reagent(s) used in the analysis

Lower detection limits are the lowest quantity of a substance that can be distinguished from the absence of that substance (in the case of upper detection limits; the highest quantity of a substance that can be reliably measured). When a result is below a detection limit, its exact amount cannot be determined, but it is still important to report this result as it conveys that the substance is at a low concentration, or not present at all. If left out from a dataset, only values over the detection limit will be displayed, skewing any statistical analyses of that substance.

When a result is below or above a detection limit, the ResultValue field is left blank. The result detection condition (Table 5) and quantitation limit measure, unit and type must be inputted to adequately describe the result.

Table 5: Detection condition terms used in DataStream that are applicable to metal-series data.

This document was prepared in collaboration with Dr. Megan Thompson of Thompson Aquatic Consulting.

Gleyzes C., Tellier S., Astruc M. 2002. Fractionation studies of trace elements in contaminated soils and sediments: a review of sequential extraction procedures. TrAC Trends in Analytical Chemistry 21(7): 451-467.

Tessier A., Campbell P.G.C., Bisson M. 1979. Sequential extraction procedure for the speciation of particulate trace metals. Analytical Chemistry 51 (7): 844–851.

USEPA, 2018. Best Practices for Submitting Metals Data to the Water Quality eXchange (WQX). https://www.epa.gov/waterdata/wqx-metals-best-practices-guide.