Ecological monitoring of the Dnepr reservoir north of Kiev with the focus on drinking water issues


M. Hoffmann

Centre for International Migration and Development, Germany,

Consultant in the Ministry of Environmental Protection and Nuclear Safety of Ukraine

V.I. Rakov

Main State Ecological Inspectorate of the Ministry of Environmental Protection and Nuclear Safety of Ukraine

N.S. Kravets

Main State Ecological Inspectorate of the Ministry of Environmental Protection and Nuclear Safety of Ukraine

O.A. Kryzanska

Main State Ecological Inspectorate of the Ministry of Environmental Protection and Nuclear Safety of Ukraine

ABSTRACT: Precondition for a sustainable water management and an ecologically sustainable planning is an excellent understanding and regular observation of the ecological behaviour of river systems including the catchment area. In Ukraine, Dnipro monitoring has mainly been limited to chemical water controls of maximum permissible concentrations, not really including ecological aspects. This is also due to limitations of personnel, equipment and financial reasons. The study presented herewith should give some hints as how to proceed without big expenditures.

The proposed methods allow to collect information about the trophical state of the river reservoirs, to differentiate between various types or origins of organic loads and to plan future monitoring of water designed for drinking water purposes.


The river Dnipro and its tributaries drain a catchment area of more than
500 000 km2, situated within three countries. Belarus and Russia in the North contribute more than 60 % of the river water, Ukraine less than 40 %. It is the third biggest river in Europe. For Ukraine, it has eminent economical, political, and social importance, like the Volga in Russia and the Rhine in Germany. Without any doubt, its biggest importance is the supply of raw and drinking water. More than 70 % of Ukrainian drinking water are taken from the Dnipro supplying about 30 million people. Taking this into consideration, it turns out that the protection of the river water quality and likewise of its ecological function must have the highest priority today as well as in the future. In fact, Ukraine and several countries, supporting Ukraine in environmental matter, take pains to conserve or even to improve the ecological state of the river carrying out investigation projects, restoration and action programs including financial investments.

Within this context, a bigger number of impacts can be summarised as follows:

a) impacts related to various types of land uses, including deforestation, agriculture, erosion, use of fertilisers and pesticides

b) point source pollution like industrial discharges, sewage treatment plants and surface runoff

c) hydrological measures, especially the construction of water reservoirs

In the northern part of the river basin, this is 2/3 of the catchment area, the influence of agriculture is dominant, while in the southern part industrial impacts are crucial. The 6 storage basins of the Dnipro follow each other with only short distances in between. The Kiev reservoir begins already a short distance after the border (Belarus/Ukraine), not far from the well known town Chernobyl. The building of dams has increased the flow time about 30 fold. Nutrients deriving from agriculture and damming lead, as one can expect, to such consequences as eutrophication, accumulation of oxygen consuming matter and toxic substances in the sediment.

Furthermore, the water quality is obviously influenced by swamps and wetlands. Many of them are accumulated in the north-western part of the river basin and spend their water into the Pripjat, the main tributary of the Dnipro (Pripjat swamps). Humic matter in connection with manganese and iron give the water a well visible red-brown colour, especially during Spring. Correspondingly, they influence the physical characteristics as light and vertical distribution of temperature, the chemical properties (improvement of the photochemical degradation) and finally also the biological relations. Further, humic matter leads to problems in the water works (Hoffmann 1995) and constitutes precursors for chlororganic compounds.

The river Dnipro has been investigated by various national and foreign organisations, mainly from Canada (Vasenko 1998) and USA. The bigger part of these studies has been limited on the Dnipro section between the reservoir situated south of Kiev and the Black Sea. The main aim was the documentation of anthropogenic loads through punctual discharges from industry and households and some ecological consequences. The Kiev reservoir (north of Kiev) is sampled by a State control organ, but usually no further difference is made between the various organic loads.

Already before the start of the investigation, it turned out that the composition of the Dnipro water is submitted to significant and not predictable changes all over the year. Thus, in Winter 98/99, but not in the Winter 99/2000, an increased turbidity, and a fish kill occurred under the ice cover. The use of Dnipro water for drinking water purposes was not possible any more for some time. Cyanobacteria blooms too, do not develop every year and not always at the same time. Therefore the investigation program had to combine various aspects. Alteration of water quality all over the year focussing on the concentration of organic matter which lead to the formation of chlororganic compounds and other problems in the water works, had to be studied. Within this frame work, type and origin of the organic load needs also be clarified because their potential to form chlororganic substances later on during the treatment process in the water work can be different. The investigations have been carried out at the southern part of the Kiev reservoir, shown in figure 1. For technical reasons, on-site measurements and sampling have been limited on 4 places near the banks (west, south and east bank), where the water seemed well connected to the main current. Another site was a short distance downstream of the dam near the Kiev water work.

fig. 1

Figure 1. Dnipro catchment area and position of the Kiev reservoir (big arrow) in northern Ukraine

The river water quality monitoring was completed by repeated investigations of Kiev tap water focussing on organic and chlororganic compounds. Final aim of these investigation series was to document the contamination of drinking water, to work out the relation between raw and drinking water quality and thus contribute to the decision making process for future drinking water supply strategies.

Another important point of view was the question of the technical realisation and the possibility of a co-ordinated follow up program by state laboratories with moderate equipment which already exist within the basin area.

The program started in March 99. At the beginning, the sampling sites have been visited twice a month, later monthly. A more frequent sampling, especially during the Summer, is planned to be organised.

Electrochemical measurements of temperature (WT), oxygen (O2), pH, and electrical conductivity (EC) have been performed in situ or on-site. The organic matter has been quantified with several different techniques to gather additional information. At first, the total biochemical oxygen demand in 5 days (BOD5) and the chemical oxygen demand (COD) have been measured by the usual standard methods. In the laboratory, the absorption spectra of the samples have been recorded between 205 and 436 nm with an UV/VIS photometer. This was possible because the samples were visually clean in nearly all cases. This was controlled by the spectral absorption coefficient at 436 nm (SAC436; German standard method DIN 38404-C1 in: DEV 1993) and by turbidity measurements (NTU; standard method ISO 7027, DEV 1993).

The organic load has also been estimated by measurements of the spectral absorption coefficient at 254 nm (SAC254). Following the standard method (DIN 38404-C3 in: DEV 1993), results are given in 1/m. Especially aromatic compounds, like humic acids, as well as molecules with double bounds (C=O, C=N; N=O) absorb at 254 nm. From SAC254 it is possible to roughly estimate the concentration of other parameters like e.g. TOC and COD. COD figures (dichromate method) are usually slightly higher than SAC254. The factor found in (bigger) German rivers was about 1.2 (Hoffmann, unpublished data).. A relation to the KMnO4-consumption was given by Mueller et al. (1984). Iron does not disturb the SAC254 measurement if the quotient mol Fe/mol DOC < 20 % (Buffle et al. 1982a).

Humic matter and other molecules with double bonds also show a stronger fluorescence. This ability was used to better describe the organic matter as a whole as it was proposed e.g. by Banoub .(already in 1973), Buffle et al (1982 b), Stabel et al. (1983), and Rostan et al. (1986). For technical reasons, the exciting light wave length was set to between 300nm to 340nm (maximum 317nm). Emission was measured using a filter for wave lengths of between 380 and 445 nm (maximum 405nm). The DOC of the sample should approximately be known or estimated and lie under 10 mg/L. Iron does not disturb the measurement if the molar quotient Fe/DOC < 10 %. If higher, a mistake up to 20 % must be taken into account (Buffle et al. 1982a). The fluorometer (firm Lumex, St. Petersburg) was calibrated using a 0.2 mg/L potassium salicylate standard. For this solution, fluorescence (F317-405) was set to 1. As for SAC254, results have also been multiplied with 100 (corresponding 1/m).

Parallel measurements of fluorescence and SAC254 have been carried out in test series of raw and drinking water (Dnipro water and Kiev tap water) already during the year 1998.

The dynamism of the phytoplankton development has been described by two methods. As indicator for the biomass (standing crop), measurements of chlorophyll a have been used as described in APHA (1992). Additionally, the photosynthetic activity has been investigated as oxygen production under laboratory standard conditions (OPL, light/dark bottles, 20 oC, 1300 lux, 24 hours (DIN 38412-L14 in: DEV 1993).

The total amount (sum) of chlororganics was measured with an AOX-analyser (firm Abimed). The samples were prepared and, for 2 hours or more, shaken following the German standard method DIN 38409-H 14 (DEV 1993) and the firm manual.

As it is common for anthropogenic loads, the analytical results mostly did not show normal Gauss distribution. For the calculation of correlation, a distribution free method (Spearman’s correlation coefficient) had to be used.

An important reason for the start of the monitoring program has been strong water colourings and turbidity. They could be traced back to a mixture of colloids and particles coming up from an increased re-solution of substances like manganese, organic matter, bacteria, ammonia etc. from the sediment because of an oxygen deficit. For the quantification of this phenomenon, measurements of colourings and turbidity (SAC436 and NTU) have been carried out and compared. The SAC436 measurement also covers filterable particular and colloidal substances and, on the other hand, NTU also includes solved compounds. Accordingly, the two parameters give similar results and show a strong significant correlation.

Another reason for turbidity and colourings, coming up during the summer months, is the phytoplankton biomass. Spearman’s correlation coefficient r for NTU/chlorophyll a was found to be 0.52 (p=0.01).

Already at the beginning of April, a short time after the ice melting, a strong phytoplankton development could be registered. The phytoplankton density changed during the year, obviously in dependence from weather and light conditions (see figure 2). Differences could also be registered between the east- and westbank, and in dependence from the wind direction.

Figure 2. Results of the monitoring of the photosynthetic activity (OPL) at five different places

Cyanobacteria came up mainly during the Summer, diatoms developed even under the ice cover in December. More accurate statements about the plankton distribution in the whole reservoir could only be produced by way of more time consuming investigations. More representative and relevant for the water work are the samplings downstream of the dam. This would however be to late for any prognosis or recommendations.

Which of those methods already mentioned in chapter 3 are appropriate for a minimum routine program, should be worked out through statistical evaluations. The question of which method fits best for the description of the phytoplankton development, must be answered in dependence of the investigation issue. The here presented methods are quick and not complicated in the practical handling. The quotient chlorophyll a/OPL (the "photosynthetic capacity") is subject to significant alterations as can be concluded from figure 3.

Figure 3. Relation between results from chlorophyll a and OPL

For the water works, identification of plankton is not less important than quantification. Especially Cyanobacteria are known to cause the blocking of filters and release undesirable or toxic substances into the water. Type and concentration of organic compounds can be described by the four parameters COD, BOD5, SAC254 and F317-405. COD and F317-405 are positively correlated with SAC254 The relation between COD and SAC254 for the river Dnipro can be seen in figure 4.


Figure 4. Comparison of SAC254 (UV absorption) and COD

SAC254 mainly represents humic matter while BOD5 (and the BOD5/COD quotient) is more influenced by phytoplankton (chlorophyll a and OPL); see figure 5.


Figure 5.Influence of phytoplankton (OPL) on BOD5

Figure 6. Relation between the quotient SAC254/COD and COD

Therefore the quotient SAC254/COD and SAC254/BOD5 give a fairly good idea about the type of organic load (see figure 6 and 7). Big figures can be expected when humic matter is more important than phytoplankton.

Figure 7. Alteration of the SAC254/BOD quotient with increasing BOD

Additional measurements of the fluorescence represent valuable data to better characterise the humic matter. The quotient SAC254/F317-405 from raw water and drinking water (Dnipro) was always similar and different from that of ground water. Humic and fulvic acids in ground water consist in smaller molecules which are known to show a better fluorescence (figure 8).

Figure 8. Correlation between F317-405 (fluorescence) and SAC254 (UV absorption) of various types of water (from Hoffmann 1999)

The weather condition do not only influence the phytoplankton distribution. During rainy periods, a bigger amount of water comes from surface runoff and drainage of arable land (agriculture) into the Dnipro leading to higher electrical conductivity and smaller organic loads. On the other hand, smaller conductivity and high SAC254 figures indicate that a bigger amount of water is released from swamps and wetlands (see figure 9).

Knowledge of SAC254 in combination with some other parameters are also useful for the water treatment process, e.g. for the application of the ”enhanced coagulation” as described by Edzwald 1994. The necessity to improve the water treatment process in the water works has been documented since 1993 (Hoffmann 1994, Hoffmann et al. 1999). Mean concentration of the sum of chlororganic substances (AOX) in raw water has been found to be 69 +/-26 ug/L. After the disinfecting (before and after treatment), high concentrations of chlororganics up to about 1000 ug/L (maximum values!) could be detected during Summer over many years.

fig. 9

Figure 9. Relation between SAC254 (UV absorption) and EC (el. conductivity)

Recently, situation in Kiev has shifted to improvements because of various technical measures and experiments. Therefore it was not possible to establish a reliable relation between the concentration and/or type of organic matter and concentration of AOX as it has been planned. In principle, this should be possible, as could be shown by SINGER and CHANG in 1989. The authors calculated the formation potential of trihalomethanes ("THMFP") from SAC254. However, they also pointed out that the equation given in their paper, should not be transmitted to water from other sources without further calibration.


The results of the research show that it is possible to gather information about the type and origin of organic matter through the application of the methods as they were proposed here. For instances, the analyses could be used to show that the basic load of organic compounds in the Dnipro originate from swamps upstream. Appearance of phytoplankton must be expected between April and December. Its concentration and distribution can change within less than one month. Monthly sampling turned out not to be sufficient for a reliable monitoring.

Phytoplankton, humic matter and other organic compounds lead to the formation of chlororganics after disinfection of raw and drinking water. It is therefore recommended to use the mentioned methods for quality controls as well as for improvements of the water treatment process. To point out the importance of the whole programme for future water resources management, results of AOX analyses have been mentioned. Correspondingly, it is also advised to introduce the AOX method in Ukraine. This would allow to investigate the concentration of chlororganic substances in drinking water and synchronously the organic load of raw water within a short time.

The bigger part of the methods as proposed here, has still not been certified by the State organisation for standards (Gosstandard). This is, however, necessary to promote an ecologically more informative monitoring.

In spite of possible water quality improvements in the future, efforts should be made to increase the amount of ground water as a raw water source instead of using surface water.


The monitoring program described here was performed for the Main State Ecological Inspectorate of the Ministry of Environmental Protection and Nuclear Safety of Ukraine in Kiev. For analytical support and permanent valuable co-operation, we wish to thank G.B. BABICH, and E.A. KALITINSKA. The work in Ukraine of the first mentioned author is funded by the Centre for International Migration and Development (CIM, Frankfurt/M, Germany) and by the Agency for Technical Cooperation (GTZ), Eschborn, Germany.


American Public Health Association (APHA), 1992. Standard Methods of the Examination of Water and Wastewater - 18th edition, pp.10-17 - 10-22

Banoub, M. W., 1973. Ultra violet absorption as a measure in natural waters. - Arch. Hydrobiol. 71, pp. 159 - 165

Buffle, J., P. Deladoey, J. Zumstein, and W. Haerdi, 1982a. Analysis and characterisation of natural organic matters in freshwaters, I. Study of analytical techniques, Schweiz. Z. Hydrol., 44, pp.325 – 362

Buffle, J., and P. Deladoey, 1982b. Analysis and characterisation of natural organic matters in freshwaters, II. Comparison of the properties of waters of various origins and their annual trend. - Schweiz. Z. Hydrol., 44, pp. 363 – 391

Deutsche Einheitsverfahren zur Wasser-, Abwasser- und Schlammuntersuchung (DEV), 1993. Beuth Verl. Berlin, (German)

Edzwald, J.K., 1994. Coagulation Concepts for Removal of Total Organic Carbon, Proc. of the American Water Works Association (AWWA WQTC), Nov 6-10, pp. 1465-1481

Hoffmann, M., and V. P. Mykhaylenko, 1994. Chlorination of Drinking Water and its Effects in the Ukrainian Capital Kiev. - Chemistry and Technology of Water, 16 (5), pp.472 – 479 (Russian)

Hoffmann, M., 1995. ber Herkunft und Vorkommen toxischer Stoffe im Trinkwasser ukrainischer Grostdte. - gwf Wasser/Abwasser, 136 (2), pp. 85 – 90, (German)

Hoffmann, M., and A. A. Galagan, 1999. Hydrochemical Investigations on the Occurrence of Toxicological Relevant Matter in Rivers and Drinking Water, Chemistry and Technology of Water,. 20 (2), pp.154 – 167 (Russian)

Htter, L.A. 1990. Wasser und Wasseruntersuchung - 4. Aufl., Verl. Salle + Sauerlnder, pp. 287-288 (German)

Mueller, K. U., and F. Randow, 1984. Lysimeteruntersuchungen zur Auswaschung organischer Verbindungen nach Glledngung und ihre methodische Bestimmung. - Wiss. Z. d. Wilh.-Pieck-Universitt Rostock, 33 (5), pp. 13-14, (German)

Rostan, J.C., and C. Mouvet, 1986. Apports de la spectrophotometrie ultra-violette et de la matire organique dissoute, l’tude du fonctionnement. - Sciences de l’eau, 5 (1), pp. 9 – 28, (French)

Singer, C.P., and S. D. Chang, 1989. Correlation between Trihalomethanes and Total Organic Halides formed during Water Treatment, Journal.Am.Wat. Works Assoc. 81(8), pp.61-65

Stabel, H.-H., and H. Schweder, 1983. Fluorescenzspektrometrische Charakterisierung der gelsten organischen Substanz eines Fliegewsser-Seensystems in SW Irland, Vom Wasser; 62, pp. 329-333, (German)

Vasenko, O.G., 1998. Environmental Situation in the Lower Dnipro River Basin, Water Qual. Res. J. Canada 33 (4), pp. 457-487

 September, 2000