It is crucial to remember that the condition of the samples is of utmost importance to get a good result in a water analysis. The water sampling should be done with great care and sampling points as well as time of sampling should be planned carefully.
When sampling a watercourse, a sampling station located where a road crosses the watercourse can be practical. The sample should then be taken upstreams from the bridge.
Sampling from lakes
In lakes, the water conditions vary dependent on depth. One factor are pycnoclines, e.g. thermocline and halocline. Water above and below a pycnocline like thermocline is a holdback for mixing. For this reason, it is important to know if a sample was taken: above, in or below the pycnocline, how well developed the pycnocline is, and on what depth it asperse. This is important information when interpreting the data. If there is only opportunity to take one sample, its recommended to take it in the middle of the lake at two meters depth. To get a fairly representative sample of a lake, the sample can be taken with a bottle directly from the outlet.
Sampling at sea usually requires a boat and thus can become expensive. Everything from a small rowing boat for sampling near the shore, to larger research vessels for the open sea might be required. Bridges, jetties, ferries, or solid ice are some examples of other platforms from which sampling in the sea can be done. If there is a possibility to visit several sampling stations, the samples should be taken along a transect beginning closest to shore and then out towards the open sea. The samples should be from both the surface and deeper layers, i.e. both above and below any possible pycnocline.
Special water samplers, such as Ruttner samplers, are usually required in order to get samples from great or specific depths. These usually contain a thermometer, which can tell if the sample is taken above or below a possible pycnocline.
There are some handling demands that apply to bacteriological sampling: bottles and other equipment must be sterile, and actions must be taken to minimize the risk for infections. A common procedure is to lower a glass bottle into the water with the help of a sterilized steel wire wrapped around the bottleneck.
Stream water collector
An arm extension, a stick or telescopic rod,
with a jar taped to the end (45° angle),
can be used to ease the sampling in small streams.
Water for analysis in laboratories is brought back in well-filled bottles made from e.g. colourless polythene. Avoid any unnecessary exposure to air when filling the bottles. Rinse the bottles with sample water and fill them to the brim. In samples for oxygen and pH analysis no air bubbles can be allowed on the walls of the bottle.
Store the bottles cool and dark during the transport to, and in, the lab and perform the analysis as soon as possible after sampling. Some analysis results are highly dependent on a short timespan between sampling and analysis. It is of great importance that sampling bottles are clean and free from metal and salt residues. If detergent is used, this has to be of laboratory type, phosphate- and acid-free. It is preferred if everything that will come in contact with the water is washed in acid, e.g. 0,1 M hydrochloric acid. The bottles shall be rinsed well with distilled / deionized water.
If the sampled water is to be used for analysis with atom absorption and other highly sensitive methods, a special cleaning is required. Any metallic parts of the water collector must be covered by plastic if the sample is to be tested for heavy metals later on.
Conservation of samples
Some samples, e.g. those for metal analysis, are to be conserved in connection to the sampling. Iron, manganese, aluminium, phosphate-phosphorus and total phosphorus is conserved with 1 ml sulphuric acid (4 mol/l) per 100 lm sample. Samples that shall be analysed for metals with atom absorption technique are conserved with nitric acid, 1 ml (7 mol/l) per 100 ml sample for flame analysis or 0,5 ml (conc.) per 100 ml sample for analysis in oven. Conservation of samples for analysis of nutrients is not recommended.
Some variables can be measured through lowering a measurement cell to various depths. The variable in situ (on site) is thereby measured. Data can be stored directly in the measurement cell or be sent through a cable to a reading instrument above the surface. Salinity and oxygen are examples of variables that can be measured in situ.
When sampling a lake, it is not only the depth sampled that is important. A lake can be divided into different water masses, like Rammsjön in the figure above.
If there is a big difference between the inflow water from the two streams, the upper part of the lake will have a different water quality than the lower part.