The purpose of sampling and chemical analysis of precipitation in the EMEP network is generally to give an accurate indication of precipitation chemical composition, which can be used to derive deposition by scaling with precipitation amounts, both on short-term (day-month) and on long-term bases.
In connection with the determination of transboundary fluxes and deposition of air pollutants, the concentrations of sulphate, ammonium and nitrate in precipitation are particularly important. However, determination of one or more of the sea-salt constituents (Na, Cl, Mg) is also necessary in order to determine the fraction of sulphate concentration which is due to marine sea-spray aerosols; and determination of the base cations Ca, K, and Mg is desirable in order to give an indication of the large-scale deposition of bases which is needed in connection with the determination of critical loads.
Finally, pH and conductivity
should also be determined in order to give an indication of the
overall composition of the samples, and to check the consistency
of the chemical analyses.
Precipitation is collected in a
vessel, with a defined horizontal opening. The collecting vessel
must be constructed from a material, which does not alter the
chemical composition of the sample, and shall give a reliable
measure of the amount of precipitation on a daily basis. The
concentration of the major anions and cations are determined by
chemical analysis.
In order for the measurements to be useful for validation of models of long-range transport and deposition of air pollutants the site for precipitation collection should be chosen, and the collection of rain and snow for analyses should be made in such a way that the concentrations are representative of rainfall composition over a larger area. For this purpose, the following requirements have been worked out:
1. The annual precipitation amount at the site, as measured by an ordinary meteorological precipitation gauge, should not differ markedly from the precipitation amounts at adjacent sites in the national precipitation network, and the daily precipitation amounts should also be correlated with those from the adjacent sites.
2. The location of the sampler should conform to WMO site requirements for precipitation gauges (WMO, 1971). There should be no obstacles, such as trees, above 30o from the rim of the precipitation collector, and buildings, hedges, or topographical features which may give rise to updraughts or downdraughts should be avoided. Consideration of the prevailing wind directions during precipitation events is recommended in connection with locating the sampler.
3. Of particular concern is the sedimentation of soil dust particles from the immediate surroundings. Gravel roads, farmyards, and tilled agricultural fields in the near surroundings within a distance of 100 m to 1 km should be avoided. Other potential local contamination sources include local residential heating with wood, peat or coal. Potassium is an indicator of such contamination. Local high ammonia concentrations from farming activities should also be avoided.
Supply of electricity is necessary for the operation of wet-only precipitation samplers. For the operation of the sampling site a small room is needed to store samples, equipment, and documents. This must be equipped with a refrigerator for the storage of collected precipitation samples.
The sampling equipment consists in principle of a funnel and a receiving vessel. In order for the sample not to be contaminated from the ground during heavy rain, the rim of the funnel should be positioned 1.52 m above the ground level. It is recommended that the sampler be further protected from sedimentation of dust and adsorption of gases during dry periods by an automatic lid, which opens after activation of a precipitation sensor. The precipitation sensor is usually based on measuring the electrical conductivity between a pair of gold-plated electrodes on a suitable non-conducting surface. The sensor is electrically heated to a temperature of 12 degrees above the ambient temperature so that the water film evaporates after the precipitation event. The sensitivity of the sensor is important, a precipitation amount of 0.05 mm/h should be sufficient for the lid opening mechanism to be activated.
Precipitation collectors are commercially available and a list of instruments and manufacturers' addresses is given below (Table 3.1.1). In selecting one of these, reference should be made to available field test results (e.g. Winkler et al., 1989; Granat et al., 1993), and the climatic conditions at the site should also be considered.
Bulk samplers are recommended only if it can be shown that the contamination by dry deposition of dust and gases e.g. ammonia is negligible, and during periods when the precipitation is mainly in the form of snow. Wet-only samplers are unsuitable for collection of snow, because of generally poor aerodynamic designs, and because heating of the funnel to melt the snow may cause serious evaporation and concentration of the sample. The response of the conductivity sensors to dry snow is also poor.
Table 3.1.1: Commercially available wet-only collectors.
NSA181 | DRA-92 |
MISU | WADOS |
Firma Eigenbrodt Königsmor/Kr.Harburg D-21255 Germany. http://www.eigenbrodt.de |
Digitel Elektronik AG CH-8604 Hegnau Switzerland http://www.digitel-ag.ch/ |
Department
of Meteorology, Stockholm University, S-106 91, Stockholm, Sweden. http://www.misu.su.se |
Kroneis GmbH A-1190 Wien Austria http://www.kroneis.at |
![]() |
![]() |
![]() |
![]() |
All materials that come in contact
with the sample must be chemically inert. Polyethylene,
tetrafluoroethylene and tetrafluoroethylene-fluorinated ethylpropylene copolymer are generally recommended because of
their excellent chemical properties. The mechanical properties of
these materials must be taken into account in the construction of
samplers, however. Polyethylene may become brittle when exposed
to sunlight, and should be replaced after 1 year of use involving
exposure to sunlight. Borosilicate glass should be properly
acid-washed and rinsed in deionized water prior to use, but the
use of glass is not generally recommended. Soft glass will
contaminate the sample with alkali and alkaline earth cations.
Metals and artificial materials with unknown chemical properties
or composition should be avoided. If such materials have to be
used in joints or in other constructional details, boil a sample
of the material in deionized water and analyze the water
afterwards as a precipitation sample!
The construction principles for precipitation gauges are relatively simple. The sampler should not be too large or bulky, because this will obstruct the air flow around the sampler. On the other hand, the diameter of the collector must be large enough to provide samples large enough for chemical analysis. In practice, a diameter of 20 cm is sufficient. If a funnel is used, there should be a vertical section of at least 5 cm height.
For a general review of errors in the collection of precipitation in precipitation gauges, reference is made to Sevruk (1989). There are 4 sources of error if the sampler is equipped with a sensor-activated automatic lid:
(1) aerodynamic error, when the gauge fails to catch the same amount of precipitation which falls on the surface,
(2) evaporation errors, when part of the precipitation evaporates after the precipitation event, and before the amount is collected or measured, and
(3) the wetting error, which occurs because not all of the precipitation is transferred to the measuring cylinder,
(4) failure of the lid to open in situations with light precipitation or snow.
The aerodynamic error is reduced by proper design of the collector, and by choosing the sampling site carefully. It cannot be entirely eliminated and may be very serious at windy sites, and particularly for snow. Use of a windshield (Nipher or Wyoming) may be necessary for sites where a large fraction of the precipitation is in the form of snow.
However, at many sites the collected amounts of precipitation will be a good measure of the precipitation amounts even if no particular measures have been taken. Consultation and co-operation with meteorological services running precipitation gauge networks is strongly recommended when selecting sites and precipitation collecting equipment.
The evaporation effect is reduced if an automatic closing lid is used, but the lid must form an airtight seal with the rim of the collector. For bulk collectors of the funnel and bottle type, diffusion through the funnel stem reduces the evaporation effect.
The wetting error is due to the liquid film on the inside of the collector. The equivalent volume of this liquid film is related to the internal surface of the collector, and may be determined experimentally, for example by weighing the collector in dry condition, spraying with water, "emptying", and weighing again.
It is not unusual to find a wetting error corresponding to 0.2 mm of precipitation. Winkler et al. (1989) have measured the wetting film on several commercially available precipitation collectors.
Evaporation changes are particularly serious, since this may result in a significant concentration of the sample. Electrical heating of the precipitation collector in order to melt precipitation in the form of snow is therefore not recommended. It is acceptable, however, to apply electrical heating when the lid is firmly closed.
In order to obtain a more accurate measure of the precipitation amount, WMO GAW has equipped their sites with a rain gauge in addition to the wet-only collector. This improves the measurements of deposition, and will, provided that the Steering Body agrees, be implemented also for EMEP sites as one step in a harmonization process.
When precipitation is in the form of snow, it is advisable to use a special snow collector, in the form of an open polyethylene cylinder of diameter 20 cm. The height of the cylinder should be at least twice the diameter to prevent "blow-out". The snow collector should be equipped with a tight-fitting polyethylene lid, which is put on when the collector and sample is brought indoors for the sample to melt.
Proper design, construction and maintenance of the sampling equipment is essential in order to avoid serious errors because of poor performance of the precipitation sensor and automatic lid system. The sensor should be designed with a response which will cause the lid to open when the precipitation intensity exceeds 0.05 mm/h.
Additional equipment at the sampling site will include:
Samples are collected on a daily basis, at the same time as used in the official precipitation measurement network. Usually this will be at 0800 local time. If daylight savings time (summer time) has been introduced, samples should be collected according to the "normal" time. The daily sample collection involves transfer of the sample to a sample storage and transport bottle, measuring the sample volume, and cleaning of the equipment which has been used. The exact procedure will vary according to the equipment used at the sites. A detailed, written standard operating procedure must be worked out for each site and should be readily available at the site, in the national language of the operator. As an example, the procedure could consist of the following steps:
1. Collect the equipment needed for change of samples. Label the storage and transport bottle with station code and name, and start and end of the sampling period.
2. If there is any chance for the operator to touch the inside of the collecting funnel, disposable polyethylene gloves should be put on.
3. Exchange the collection bottle in the precipitation sample collector and put on a screw-stopper. Check that the collection equipment functions correctly by putting a drop of water on the precipitation sensor. Examine the collector funnel for visible contamination such as insects, leaves or tree-needles, organic debris. If this is found, remove the contamination and rinse with distilled water. If a bulk collector is used, the collecting funnel should be rinsed with distilled water every day. After the distilled water has drained off, put on the new collection bottle.
4. Take the collection bottle indoors to the room assigned to function as the sampling laboratory.
5. Weigh the bottle, transfer a suitable aliquot (50-100ml) to the labelled storage and transport bottle. (Alternatively, measure the volume in a graduated cylinder. Use a large cylinder (0-250 ml) for large samples, and a small (0-25 ml) cylinder for small samples).
6. Put the storage and transport bottle in the refrigerator until it can be sent to the laboratory for chemical analysis.
7. Pour out the remainder of the sample, rinse with distilled water and place the collecting bottle upside down in a clean place to dry. Also rinse the graduated cylinders.
8. Take off and discard the disposable plastic gloves.
9. Fill in the field sample registration form, and take time to record usual and unusual events which may have influenced the sampling. Examples are given below (these should be elaborated for each site, because of the different conditions):
Most of the major ions in precipitation samples may be determined by ion chromatography, which is the generally recommended method for anions such as chloride, nitrate and sulphate. Table 3.1.2 gives a list of alternative recommended methods, with reference to more detailed descriptions and procedures in Section 4. It is not recommended to filter samples.
Table 3.1.2: Recommended and alternative methods for chemical analysis of precipitation within EMEP.
METHODS |
||
Component or parameter |
Recommended methods |
Alternative |
Conductivity |
Conductivity cell and resistance bridge |
|
Hydrogen ion (H+) |
Potentiometry (glass electrode) pH<5.0 |
Titration |
Ammonium ion (NH4+) |
Ion chromatography |
Spectrophotometry (indophenol blue colour reaction) |
Sodium ion (Na+) |
Atomic absorption spectrophotometry (AAS) |
Ion chromatography |
Potassium ion (K+) |
AAS |
Ion chromatography |
Magnesium ion (Mg2+) |
AAS |
Ion chromatography |
Calcium (Ca2+) |
AAS |
Ion chromatography |
Sulphate ion (SO42-) |
Ion chromatography |
|
Nitrate ion (NO3-) |
Ion chromatography |
Reduction to nitrite and diazotation |
Chloride ion (Cl-) |
Ion chromatography |
Displacement of SCN- in Hg (SCN)42-, determination of coloured Fe(SCN) complex. |
Bicarbonate ion (HCO3-) |
Titration |
|
Formate ion (HCOO-) |
Ion chromatography |
|
Acetate ion (CH3COO-) |
Ion chromatography |
The last three anions are not part
of the ordinary EMEP measurement programme. They are included
here, however, because they are found in precipitation samples in
concentrations comparable to some of the other ions, and may be
necessary to explain the ion balance and measured conductivities,
particularly for samples with pH above 5. Note that most of the
components can be determined by ion chromatography, which is
strongly recommended for the anions sulphate, nitrate and
chloride. However, ion chromatography holds no advantages over
conventional methods when it comes to determination of ammonia
and base cations.
The amount of precipitation is to be calculated from the collected sample volume, simply by dividing by the area of the sampling orifice. No corrections are to be made for sampling errors, such as undercatch, evaporation or the part of the sample remaining in the collector because of the wetting effect. An assessment of these errors should be performed and be available.
Additionally, the amount of precipitation measured by rain gauge should be reported to the CCC.
Conductivity and pH is reported in
µS/cm and in pH units, respectively. All other parameters are
reported as elemental concentrations in mg/litre. Note especially
that nitrate, ammonium and sulphate concentrations are to be
given as equivalent weight concentrations of nitrogen and
sulphur. Table 3.1.3 gives reporting units and conversion
factors.
Table 3.1.3: Units and conversion factors.
Ion |
Reporting form | Mol /mg |
Sulphate (SO42-) | mg S/litre | 31.19 . 10-6 |
Nitrate (NO3-) | mg N/litre | 71.39 . 10-6 |
Chloride | mg Cl/litre | 28.21 . 10-6 |
Hydrogen (H+) | (pH) | |
Ammonium (NH4+) | mg N/litre | 71.39 . 10-6 |
Sodium (Na+) | mg Na/litre | 43.50 . 10-6 |
Potassium (K+) | mg K/litre | 25.57 . 10-6 |
Magnesium (Mg2+) | mg Mg/litre | 41.13 . 10-6 |
Calcium (Ca2+) | mg Ca/litre | 24.95 . 10-6 |
Conductivity | µS/cm |
Before the results are sent to the
CCC, they should be examined for internal consistency by the
responsible laboratory. The procedure for this examination is
given under Section 6, which also contains data flags and gives
information on data reporting.
Site operation
Standard operation procedure must be available at the site, together with necessary equipment, deionized water for cleaning and rinsing, replacement parts for precipitation collector. Operators should be trained and required to carry out all necessary operations under the surveillance of an experienced analytical chemist or person responsible for the quality control. Operators should also be instructed on how to fill in the field sample registration form with the remarks column (see above) and to use this column extensively for reporting of conditions at the site.
If bulk samplers are used the funnel and collecting vessel must be cleaned every day.
The site should be inspected at least once a year, and the operation of the site examined by the National Quality Assurance Manager.
Field blanks and control samples
In order to check on possible contamination on the site, field blank tests should be carried out at least once every month. For this purpose, 50100 ml deionized water samples are to be poured into the sample collector at the time of collection a day without precipitation, and subjected to the same procedure as an ordinary precipitation sample.
The quality of the precipitation chemistry data is strongly linked with the performance of the chemical laboratory. Control samples should be prepared, and analysed regularly as ordinary precipitation samples, in order to keep an independent check on the chemical analyses performed. Standard rainwater samples are available from NIST and BCR, and it is advised to use such samples as an external reference solution analysed only 24 times during the year, and in-laboratory prepared control samples for daily control work. The control samples should approximate the expected mean concentration level in the precipitation samples, and may be prepared from the following compounds:
(NH4)2SO4
Nitric acid
CaSO4 · 2H2O
MgSO4 · 7H2O
NaCl
KCl
Sample transportation
The transportation time should be as short as possible and the samples contained together with freezer packs in insolated boxes.
Chemical laboratory
It is expected that the chemical laboratory is accredited under one of the laboratory accreditation systems, or is performing close to these standards, e.g. ISO 17025.
The laboratory must keep check on its performance, with respect to detection limits, precision and repeatability, by repeated analyses of control solutions of known composition, analyses of synthetic rain samples prepared by other laboratories (preferably traceable to NIST or other certified standards), and reanalysis of at least 5% of all samples.
Quality control samples are to be included in the sample series each day, and if results differ more from the expected than the targets for accuracy and precision, full reanalysis of the sample series must be carried out. Results of the analyses of control samples are to be reported to the CCC.
Data reporting and validation
The chemical analysis data should be used to check the data for consistency, by calculating the ion balance and by comparing measured and calculated electrical conductivity (Section 6).
Results from the analysed control samples should also be checked, in order to ascertain that the chemical laboratory's performance has been acceptable.
Results should also be compared with the site operator's notes, to see if untypical results are due to special activities or conditions at the site. If it is decided to reject or to correct data, the reason for the correction should be stated, and the data should be flagged. Examples of such permissible corrections may include contamination from nearby fields due to manuring or tilling, high concentrations of potassium and ammonium indicating contamination by bird droppings, a.o. Such samples should be excluded from the calculation of monthly, or yearly weighted mean concentrations.
Comparison of reported sample volumes with daily precipitation amounts from a standard meteorological rain gauge at the site is strongly advised, since this gives an independent control of the sample collection.
This assessment of the data should
be carried out on a monthly basis, as soon as chemical analysis
data are available.
The above procedures relate to normal operations of a precipitation site, assuming that there are no particular problems with the collection of the samples. This is normally the case, at least for the main constituents, at most of the EMEP sites.
By collecting samples on a daily basis, and storing the collected samples refrigerated and in the dark, it is generally hoped to avoid biodegradation of the samples. As the precipitation is now gradually becoming less acid, there may be more reason to make sure that such bio-degradation does not take place. Bacterial growth will primarily reduce the concentration of ammonium ions and organic ions.
The acidity of a sample is usually determined by the concentrations of non-marine ("excess") sulphate and nitrate, less the concentration of base cations such as ammonium, calcium, potassium and magnesium. However, if the pH is higher than 5, dissociation of dissolved carbonic acid and organic acids such as formic and acetic acid may also contribute to the observed concentration of hydrogen ions, and the equilibrium concentration of ammonium ion is a function both of the pH and the ambient concentration of gaseous ammonia. For a discussion of the chemical equilibria involving ammonia and carbon dioxide, reference is made to Charlson and Rodhe (1982). Formic and acetic acid are thought to be formed mainly by oxidation of hydrocarbons via formaldehyde and acetaldehyde, and concentrations in precipitation samples are typically 220 micro-equivalents/litre (Keene and Galloway, 1988). Other organic acids may also be present, either as a result of photochemical oxidation processes, or generally from decay of organic materials.
While the contamination of the sample by soil dust of local origin should be avoided, there is also evidence of large-scale atmospheric transport of fly ash, soil dust and desert dust. The input of base cations from such sources is large enough to be of importance in the assessment of critical loads in relation to soil acidification.
Installations of emission control devices in the latest decades have reduced the emissions of fly ash and other alkaline dust. Only total emissions in weight units are usually available, data on the chemical composition and size distributions are lacking.
Wind erosion may be a serious problem in agricultural areas, and the soil dust has sometimes been transported over quite considerable distances. Significant amounts of soil dust and alkaline material also becomes airborne in connection with agricultural tilling and harvesting operation. Burning of straw and stubble also releases alkaline material in addition to soot.
Desert dust from Sahara is frequently observed in the Mediterranean countries, occasionally also in Northern Europe. In addition to quartz and feldspar minerals, Sahara dust also contains calcite, which is readily soluble in precipitation samples.
Feldspar and clay minerals may be partly soluble in precipitation samples and contribute to the concentrations of base cations. Aluminium ions may also be present in the precipitation samples.
Determination of the main
inorganic ions and pH also allow the calculation of the ionic
balance of the samples, provided that the pH is less than 5. For
samples with higher pH, determination of the concentrations of
anions of weak acids, e.g. formate, acetate, and bicarbonate, may
be necessary in order to determine the ion balance and to explain
measured conductivities.
Charlson, R.J. and Rodhe, H. (1982) Factors controlling the acidity of natural rainwater. Nature, 295, 683-685.
Granat, L., Areskaug, H., Hovmand, M., Devenish, M., Schneider, B., Bieber, E., Marquardt, W., Reissell, A., Järvinen, O., Hanssen, J.E., and Sjöberg, K.(1992). Intercomparison of precipitation collectors for chemical analysis, HELCOM intercalibration -second stage. (Baltic Sea Environment Proceedings, 41). pp. 15-88.
Keene, W.C. and Galloway, J.N.(1988) The biogeochemical cycling of formic and acetic acid through the troposphere, an overview of our current understanding. Tellus, 40B, 322-334.
Sevruk, B., ed. (1989) Precipitation measurement. Proceedings international workshop on precipitation measurements, St. Moritz, Switzerland, 3-7 December 1989. Geneve, World Meteorological Organization (WMO/TD 328) (Instruments and observing methods. Report 48).
Winkler, P., Jobst, S., and Harder, C.(1989) Meteorologische Prüfung und Beurteiligung von Sammelgreräten für die nasse Deposition. München, Gesellschaft für Strahlen- und Umweltforschung (BTP-Bericht 1/89).
WMO (1971) Guide to meteorological instrument and observing practices. Geneva, World Meteorological Organization (WMO No. 8 TP 3).
WMO (2004) WMO/GAW Manual for the precipitation chemistry programme (report
No160)