Water sampling and analysis

Human exposure to geogenic contamination occurs primarily through consumption of contaminated water. It is therefore essential to identify contaminated water sources. From an institutional perspective, this implies national surveys that help establish i) if there is any contamination; ii) where the regions of contamination might be and iii) where mitigation activities are most urgently required. For local organisations, the survey may be limited to a region of suspected contamination, with much less technical support.

Sampling and analysis of water is a time-consuming and costly process, and planning is one of the most important steps of any field campaign. Often health symptoms provide the first indication of geogenic contamination. The first step is always to evaluate already available information, e.g. government agency reports or academic studies on water quality. Our experience shows that relevant data often exist, but sharing these data can be a problem. Next, it needs to be decided where more information is required, which water quality parameters are essential and which instrumentation for the analysis of As and F is available. Finally, the necessary preparations need to be taken before going into the field. The following sections give an overview of sampling and measuring procedures.

Basic principles

Both fluoride and arsenic are tasteless, odourless and colourless in water. The only way to detect these contaminants is through chemical analysis. If water-quality data are not already available, a field sampling campaign is necessary to find out if arsenic and fluoride concentrations are above the relevant WHO guidelines (10 µg/L for arsenic and 1.5 mg/L for fluoride) and/or national guidelines. In a first step, only a selection of water sources in areas indicated to be at risk, perhaps by the observation of fluorosis or arsenicosis symptoms in local populations, are screened. It may be possible to prioritise certain geographic areas, which are thought to be more vulnerable to geogenic contamination, for testing.

If the screening confirms elevated fluoride or arsenic levels in even a few water sources, then a more time- and resource-intensive testing of all water sources (blanket testing) should be carried out. This needs to be done because contamination levels can vary greatly over short distances. If the financial resources are available, it may be worthwhile not only to measure arsenic and/or fluoride concentrations, but to undertake a full water analysis (sum parameters, major components, minor components), as this gives a much more complete picture of water chemistry and might yield explanations for the occurrence of geogenic contamination.

Selection of measurement method

Arsenic and fluoride analyses may be carried out directly in the field using semi-quantitative or quantitative field kits. The samples may also be taken back to a laboratory for analysis. Semi-quantitative field test kits are only recommended to classify wells as above or below an acceptable limit, while quantitative measurements provide information on arsenic or fluoride concentrations. Quantitative measurements allow us to evaluate the health hazard and are essential for mitigation planning.

Field testing versus lab testing

Field test kits have the advantage of providing immediate results in the field, allowing water sources to be marked as safe or unsafe straightaway. They also allow a check to be made for alternative safe water sources in the immediate surroundings of the contaminated well. The possibility of sharing safe sources can be discussed on the spot (keeping in mind that microbial contamination may be a problem). However, field measurements are more prone to human error, as they are performed under suboptimal conditions, and often by different testers. Laboratory equipment will produce results of superior accuracy and precision to field test kits, if correctly operated and maintained by well-trained and dedicated staff. However, there are three main obstacles to the exclusive use of laboratory methods in large screening exercises (Kinniburgh and Kosmus, 2002):

• The lack of sufficient laboratories of the required quality to process large numbers of samples reliably (though a large sampling campaign might allow long-term capacity building and result in improving laboratory performance).
• The lack of management experience to organise the collection and tracking of samples and reporting of results on a large scale, resulting in the risk of results being misreported.
• Logistical problems associated with the transporting of samples from the field to the laboratory and relaying the results back to the field.

Evidence shows that well-designed and well-implemented arsenic survey programmes using field test kits can be reasonably accurate and comparable to laboratory tests (Rosenboom, 2004; Steinmaus et al., 2006; Jakariya et al., 2007, George et al., 2012; Spear et al., 2006). The same can be expected for fluoride surveys. Testing campaigns have to be carefully planned:

1. Select sampling sites, measurement method and quality-control plan
2. Train staff involved in sampling procedures, preservation and/or transportation of water samples and handling of analytical equipment
3. Prepare monitoring forms
4. Prior to each sampling trip: check and carefully pack equipment. Often forgotten: stickers and waterproof pen for labelling, spare batteries, screwdriver for opening battery case, distilled water, pipette, GPS etc.).

Accuracy and precision

Fig. 4.1 Difference between accuracy and precision

Regardless of the equipment used, sample concentrations are obtained by comparing an analytical signal to standards or known samples. While in semi-quantitative methods these may be colour charts, in quantitative methods these will be blanks (distilled or deionised water containing analyte chemicals) and known concentrations. Laboratory analyses of >20 sample batches will usually comprise a blank and standards (between 3 and 8 standards) at the beginning and end of analysis, with one blank and one standard every 10 samples. Ideally, samples will be analysed in duplicate or triplicate. In the field, the number of analytical checks may be reduced (for practical reasons) to one blank and only a few standards at each sampling location. It is therefore recommended to make quality control checks on field kit analyses and to cross-check 5–10% of the water samples with measurements made in reference laboratories (APHA, 2012). The multiple analysis of the same sample gives a mean. The precision is the scatter around the mean (UNICEF, 2008a; Fig. 4.1). If the results lie close together, the precision is said to be high. However, their accuracy is dependent on how close they are to the “true” value. The accuracy and precision of an analytical procedure will depend on a number of factors, including the skill of the analyst, the proper operation and maintenance of the equipment and the quality of reagents used. For screening, quantitative accuracy may not be essential; if the countrywide drinking-water standard for arsenic is 50 µg/L, a field test kit does not need to be able to distinguish reliably between 200 and 300 µg/L in order to identify the well as contaminated. In India and Bangladesh, arsenic surveys have used field test kits in a semi-quantitative way to classify wells as above or below the acceptable limit of 50 µg/L. The operation of field test kits is normally easy and explained well in the user manuals of commercially available products. Nevertheless, good training on the use and maintenance of field test kits is a key factor in obtaining accurate measurements. Sophisticated laboratory methods can only be installed and operated by experienced and well-educated laboratory staff. Some fluoride and arsenic tests might depend on pH or be influenced by competing ions in the water sample. It is important i) to make an in-depth study of user manuals and ii) to consult experts if necessary, to avoid such interferences.

Costs and availability

Analytical costs depend on the number of measurements planned. For instance, the capital costs for an ion-selective electrode (ISE) to measure fluoride is high, so if only few tests are carried out, the cost per sample will be high. However, if many measurements are conducted, the running costs per test are lower for the ISE method than for most of the fluoride field test kits. On the other hand, for arsenic, the costs per measurement are lower for field test kits than for laboratory analyses. Importing chemicals from abroad can be expensive and complicated, making it preferable to obtain them from a local supplier.

Health and safety

Many of the reagents required for arsenic and fluoride measurements are harmful when in contact with the eyes or skin. They have to be carefully stored during transportation, and safety equipment (gloves and glasses) needs to be worn when handling the chemicals. Children need to be kept away from the work area, and all waste must be taken away from the field and disposed of responsibly. Another issue related to arsenic field test kits is that they may expose the analyst to unsafe levels of the toxic gas, arsine. One study found that nearly half of the arsine generated during analysis escaped from the reaction vessel (Hussam et al., 1999). Newer kits are better designed, but the analyses still need to be conducted in a well-ventilated area (i.e. outdoors). The transport of reagents in the cabin or hold of an aeroplane may be prohibited. Cargo companies or postal services are an alternative. Some documentation might be necessary for customs.

Ensuring safety: It is recommended that contaminated water sources be clearly marked (e.g. A red pump spout for contaminated water sources and green spout for uncontaminated water sources (UNICEF, 2008b)), so that it is obvious to local users whether a well is contaminated and that water should not be used for drinking or cooking purposes. Appropriate colours should be determined by consultation with the local population. It is recommended to label the well with its measured As or F concentration, as well as with the date and method of analysis.

Subchapters and Handbook chapter

• Subchapters on GAP Wiki: Water sampling and analysis - Arsenic, Water sampling and analysis - Fluoride, Detailed water analyses
• Geogenic Contamination Handbook: Please find here the PDF of the complete handbook chapter "Water sampling and analysis".