Full Papers

Correlation Between Entrance Velocities, Increase in Local Hydraulic Resistances and Redox Potential of Alluvial Groundwater Sources

Milan A. Dimkić1, Milenko Pušić2

 

 

1 Jaroslav Černi Institute for the Development of Water Resources, 80 Jaroslav Černi St., 11223 Belgrade, Serbia; E-mail: This e-mail address is being protected from spambots. You need JavaScript enabled to view it , This e-mail address is being protected from spambots. You need JavaScript enabled to view it

2 University of Belgrade, Faculty for Mining and Geology, Department for Hydrogeology, 4 Djusina St., 11000 Belgrade, Serbia; E-mail: This e-mail address is being protected from spambots. You need JavaScript enabled to view it , This e-mail address is being protected from spambots. You need JavaScript enabled to view it

 

 

Abstract

Water well capacity decreases under certain conditions. This is a result of already extensively-researched mechanical clogging, but also of chemical and biochemical clogging (biofouling). In alluvial environments, deposits of trivalent iron and carbonates are often formed on well screens and in the near-well region, which increase the local hydraulic resistance (LHR) of the well. Research conducted on several locations in Serbia has shown that the LHR increase, of a biochemical nature, largely depends on the oxic state of the aquifer and groundwater. A correlation was established between LHR increase, iron concentration and redox potential of groundwater. Based on a pre-defined well clogging rate, the correlation between the groundwater redox potential, iron concentration and allowable (critical) entrance velocity to the well was quantified. It was concluded that under certain conditions it is necessary to tighten the allowable entrance velocity criterion which ensures the aquifer's filtration stability. This considerably decelerates the LHR increase, or well ageing, and is reflected in the nature of well maintenance.

Keywords: groundwater, biochemical well ageing, local hydraulic resistance (LHR), critical entrance velocities.

 

 

Introduction

Big cities (economic and administrative centers) are in many cases situated in the valleys of large rivers. Water supply to such cities is provided solely, or partially, through groundwater extraction from alluvial aquifers. The characteristics of such aquifers depend on the part of the river course in question. In general, the thickness of alluvial deposits increases from the source of the river to its mouth and the number and structure of the strata become increasingly complex. Downstream, fine-grain fractions gradually become dominant, Fig. 1.

In such environments, some of the main characteristics of water supply sources that rely on groundwater (groundwater sources) are as follows:

  • They are of the bank filtration type, the artificial recharge type (with infiltration ponds), or a combination of the two;
  • Groundwater is (in most cases) extracted by means of wells: either tube wells or radial wells;
  • In the case of bank filtration, wells are distributed along one or both riverbanks, while in the case of artificial recharge they are grouped around infiltration ponds;
  • The locations and configurations of the wells depend on the thickness and groundwater flow characteristics of the water-bearing medium (the aquifer);
  • The capacity of the groundwater source and of the individual wells is limited by the groundwater flow characteristics of the aquifer, the hydraulic link with the river, the hypsometric relationships between the riverbed, aquifer, river stage and well screen, the distance of the well from the river, and the structural characteristics of the well;
  • The presence of a semi-permeable aquifer roof ensures sound protection from surface pollution;
  • The self-purification potential of the aquifer ensures high and consistent groundwater quality;
  • Groundwater sources are often at a relatively small distance from the city and this reduces the cost of transportation of water to consumers. However, this situation sometimes creates a conflict between the interests of the city (expansion, development) and the existence of the groundwater source; and
  • Under certain conditions, well capacity decreases over time, more or less rapidly. This process is called well ageing. A correlation between entrance velocities to the well screen and well capacity was established a long time ago.

fig01

Figure 1: Change in grain sizes of an alluvial aquifer along a river course (Dimkić et al., 2011a).

 

Well ageing is generally a result of two processes: well clogging and well corrosion. They can take place in parallel or one of them may be entirely lacking. Both processes are mechanical, chemical or biochemical. Well ageing reduces well capacity at the same groundwater level maintained in the well, or the groundwater level declines at the same discharge. In most cases, the ultimate outcome of well ageing is decommissioning.

Mechanical clogging is a process by which fine particles of the aquifer matrix become entrained by the groundwater and are transported to the well, where they are unable to pass through well screen slots and, remaining there, form sediments (clogs). This reduces the active surface of the well screen and the throughput of the well.

Mechanical clogging has been directly linked to the filtration performance of the near-well region. Filtration instability occurs relatively quickly in intergranular porosity settings, where groundwater flow velocities are such that fine particles of the aquifer matrix are picked up and deposited on the well screen and in the near-well region (Istomina, 1957, Vuković, Pušić, 1992). Based on in-situ and laboratory research, many authors (Abramov, Sichardt, Kovacs, Truelsen, Gavrilko, Johnson and others) have provided instructions and criteria regarding allowable entrance velocities of groundwater to the well, which ensure the absence of this undesirable occurrence. Dimkić, Pušić (2008) reported a number of results of similar research. As defined, allowable entrance velocities are correlated with the grain-size distribution and groundwater flow characteristics of the aquifer, Vuković, Soro, (1990), Fig. 2.

The results of a review of available data on the Belgrade Groundwater Source (BGWS) indicate that a correlation can be established between the grain-size distribution of the aquifer and the redox potential of the groundwater, Fig. 3.

 

Although with some reservation, the redox potential (Eh) was adopted as the main parameter for well ageing assessments and the definition of allowable entrance velocities, keeping in mind the above-mentioned process (Dimkić et al., 2008, Petković et al., 2008, Dimkić et al., 2011c). The primary advantage of this parameter is the possibility of conducting in-situ measurements, and its disadvantage is a relative imprecision of the collected data. The data in the figure have been "evened out", by adopting representative intervals for d50 (50% of the grain-size distribution curve of the aquifer matrix) and constructing a plot based on calculated mean values.

In alluvial environments, chemical and biochemical well clogging is manifested by the formation of sediments, mostly compounds of trivalent iron, manganese and carbonates (Barbič et al., 1974, Cullimore, 1999, Dubinina, 1978, Houben, 2003, Houben, Treskatis, 2007). The development of such deposits is accompanied by an increase in local hydraulic resistances at the well screen or in the near-well region (each new well initially exhibits a local hydraulic resistance, LHR, whose magnitude depends on several factors), Dimkić et al., 2011b. Research conducted at several alluvial sites in Serbia has shown that a biochemically-induced LHR increase largely depends on the oxic state of the aquifer and groundwater. A correlation was established between LHR increase, iron concentration and redox potential of the groundwater. Based on a previously specified rate of well clogging, the correlation between the groundwater redox potential and the allowable (critical) entrance velocity to the well was quantified. It was concluded that under certain conditions, it is necessary to tighten the entrance velocity criterion which ensures the aquifer's filtration stability. This tends to substantially decelerate the rate of LHR increase, or well ageing, and is reflected in the nature of well maintenance.

 

fig02

Figure 2: Allowable entrance velocities of groundwater to the well where filtration stability is maintained, Vuković, Soro, (1990).

 

fig03

Figure 3: Correlation between the grain-size distribution of an alluvial aquifer (d50) and the groundwater redox potential (Eh) at the Belgrade Groundwater Source.

Well Ageing by Iron Clogging and Bank Filtration Stages

Fine-grain fractions of alluvial aquifers along lower courses of large rivers often include minerals that contain iron. As a result, there are massive populations of iron bacteria on such locations. The sediment created is the main cause of well ageing and well capacity decline. Such ageing occurs where the redox potential is relatively low and there is enough dissolved iron in the groundwater. Apart from natural conditions, some of the important considerations include the material of the well, its initial capacity, operating conditions and maintenance.

For large-scale development of sediment, the redox potential in the near-well region needs to be elevated (assuming there is more oxygen than in the groundwater beyond the near-well region). Then a reaction of the following type takes place:

For01(1)

This reaction is catalyzed by bacteria, which use the energy of the reaction and/or its products for their needs. According to Jurgens et al., 2010, oxidation-reduction processes proceed as follows:

for02(2)

River (surface) water, on its way to the well, undergoes several chemical transformation stages. Depending on the oxic state, changes occur that result in well ageing due to iron clogging. Four stages of groundwater flow can be distinguished, as shown in Fig. 4 (Dimkić, 2012, Dimkić et al., 2012).

 

The first stage is filtering of river water through the riverbed. The occasionally mobile bedload is generally highly oxic and characterized by extensive sorption of most organic and other dissolved substances, as well as high biochemical activity.

In the second stage filtration takes place under oxic aquifer conditions and dissolved oxygen is used for the oxidation of organic substances and volatile minerals. Of interest here is the effect of dissolved oxygen on the oxidation of volatile aquifer minerals, which contain bivalent iron, and the conversion of iron into insoluble trivalent forms.

In the third stage, which takes place at a somewhat greater distance from the river than the previous two, the oxic state of the groundwater is lower and the nature of the transformation processes different. Fe2+ might occur at this stage, which is highly soluble in water. If the aquifer is highly oxic, the third stage of groundwater transformation may be missing.

The fourth stage of bank filtration/groundwater transformation takes place in the immediate vicinity of the well, where mechanical, chemical and biochemical changes occur due to intensive processes at the aquifer/well screen interface. If the aquifer is anoxic around the well (with a somewhat higher redox potential inside the well), Fe2+ is again converted into Fe3+ at the aquifer/well screen interface.

 

fig04

Figure 4: Bank filtration stages (Dimkić et al., 2012).

 

Local Hydraulic Resistance and Entrance Velocity

The parameter generally used to quantify the efficiency of a well is its specific capacity, q:

for03(3)

where: Q – well capacity (L/s) and S - well drawdown (m). Here drawdown is defined as the difference between the static (natural, undisturbed) groundwater level and the (dynamic) water level in the well (assuming, of course, that the well is operating).

An inverse, modified quantity is referred to here as the local hydraulic resistance (loss) and abbreviated as LHR (Dimkić, Pušić, 2008, Dimkić et al., 2012):

for04(4)

where ∆S (m) is the local drawdown—difference between the piezometric head of the groundwater at the contour of the near-well region (adjacent to the well) and the water level inside the well, and v (m/s) is the average entrance velocity to the well (in practice, the discharge of the well can be used instead of v, and in the case of radial wells also the discharge of the well lateral). LHR is a parameter that can be used to quantify the extent of well clogging.

In practice, a piezometer installed in the immediate vicinity of the well, sometimes referred as the "nearby piezometer", is used to measure groundwater levels outside of the well, Fig. 5. At the same discharge, the difference between groundwater levels from the piezometer to the clogged well does not change over time and can be disregarded in ideal conditions. In practice, this parameter can be used to compute LHR without a large error.

The same assumptions are made for a radial well; the water level inside the well is equal to the water level inside its lateral.

An increase in LHR over time does not necessarily mean that well discharge will decrease. At the same discharge, the water level inside the well decreases until there is no more room in the well due to the position of the pump or other constraints.

The rate of clogging, KLHR (kinetics of local hydraulic resistance), is defined as "LHR variation in time interval ∆t":

for05(5)

 

Based on research conducted at the Belgrade Groundwater Source, the rate of clogging depends on several parameters, some of which can be used as well clogging indicators:

for06(6)

where: v – entrance velocity, Fe – iron concentration in well water, Eh – redox potential, B – function of the growth rate of bacteria in the well, Γ – function of several structural parameters (well with or without gravel pack, gravel pack characteristics, type and characteristics of screen slots) and the grain-size distribution of the aquifer.

It is apparent that KLHR can, inter alia, be considered a function of the rate of groundwater extraction, or the magnitude of the entrance velocity. The question is raised of predicting the change in local drawdown, which in the given time interval should not exceed the pre-defined, allowed value - ∆SAV. For practical reasons, the time interval ∆t is one year, such that ∆SAV is defined as the maximum allowed drawdown on account of LHR increase and serves as a criterion. Compliance with this criterion is deemed to ensure long-term service of the well:

for07(7)

where: ∆SAV – specified LHR variation during the year, expressed via increase in well drawdown (m), vperm – allowable entrance velocity (m/s), and KLHRyearLHR variation over one year. Allowable entrance velocities to the well (or well laterals, in the case of radial wells), taking the annual LHR variation as the applicable criterion, can be calculated as follows:

for08(8)

Long-term service of wells requires the above criterion, as well as the filtration stability criterion, to be fulfilled. However, compliance with these two conditions will not guarantee the full cessation of well clogging. It only reduces it to a level which ensures that the annual increase in hydraulic losses in the zone between the outer medium and the inside of the well will remain below the specified value of ∆SAV .

The case study of the Belgrade Groundwater Source revealed that at nearly all the wells vperm was much lower than estimated using "conventional" experimental formulas related to the filtration stability of the near-well region. This was especially true of anoxic/suboxic conditions, in this case 70 mV ≤ Eh ≤ 150-200 mV.

 

fig05Figure 5: Increase in local hydraulic resistance over time, tube well (t3 > t2 > t1).

Study Areas
Given that the well ageing problem is not encountered solely in connection with groundwater sources, but extends to other riparian well systems with diverse functions, wells used for different purposes and of different designs were selected for the present study. The study areas are related to three rivers: the Danube, the Sava and the Velika Morava, whose aquifers differ in terms of origin, evolution and current characteristics. Systems of wells were analyzed on five locations in alluvial sediments, Fig. 6:

  • The Sava: groundwater source that services the City of Belgrade,
  • The Velika Morava: groundwater source that services the City of Požarevac (called “Ključ”); and
  • The Danube: series of wells that are part of a drainage scheme which protects riparian lands, sections Kovin-Dubovac, Knićanin-Čenta and Veliko Gradište.

There is voluminous aquifer and groundwater information about these locations. However, the availability of necessary data and the analyzed periods were quite different. Consequently, the results of chemical analyses for the Veliko Gradište site were disregarded as field investigations and laboratory campaigns need to be intensified.

 

The Sava River

Belgrade Groundwater Source

The Belgrade Groundwater Source is in part surrounded by urban fabric. It is comprised of 99 radial wells and about 50 tube wells, located along the Sava's bank upstream from its confluence with the Danube. The length of the series of wells is about 50 km, Fig. 7.

 

The alluvial aquifer was developed through several sedimentation cycles and sequences: sandy gravel, sands of various grain sizes, and silty and clayey sediments. The thickness of the Quaternary strata is up to 25 m. With regard to the grain sizes of the sediments, two cross-sectional zones have been distinguished, Fig. 8:

  • Lower zone: coarse-grain sediments, in which radial well laterals are installed. These sediments occasionally feature clay, sandy clay and silt interbeds and lenses: and
  • Upper zone: fine-grain sediments, with poorer filtration properties.

The grain sizes of the lower zone, in which the well screens are installed, range from medium-grain sand to fine-grain gravel, Fig. 9.

The capacity of this source largely relies on the radial wells. Since its inception (the first well was built in 1953), the city's increasing water demand has been met by adding new wells. However, a continuous decline in discharge has been noted from the very beginning, Fig. 10. Rapid expansion of the source ended in the mid-1980's and, since then, it has registered a permanent capacity decline, along with a further decrease in well discharge. Periodic well regenerations (rehabilitations) increased well discharge only in the short term.

Some of the chemical parameters of the groundwater, derived from analyses of samples collected from 2005 and 2013, and aquifer grain sizes in the vicinity of the wells, are summarized in Table 1. The values are related to the entire source and reflect data on all the wells.
Figure 11 shows the results of interpretation of several groundwater parameters, average entrance velocities to the wells and d50 (50%) of the grain-size distribution, relative to selected redox potential intervals. The horizontal axis, apart from the Eh interval, shows the number of samples used in the given interval. The main characteristic of this source is a relatively low Eh, generally below 150 mV.

 

fig06
Figure 6: Study areas along the Sava, Danube and Velika Morava rivers (www.Google Earth).

 

fig07
Figure 7: Location of the Belgrade Groundwater Source along the Sava River.

 

fig08
Figure 8: Belgrade Groundwater Source: typical lithological section through the Sava riverbank.

 

fig09
Figure 9: Belgrade Groundwater Source grain-size distribution curve envelope of the setting in which radial well laterals are installed (Dimkić et al, 2007).

 

fig10
Figure 10: Average annual discharges of the wells and of the entire Belgrade Groundwater Source.

 

fig11
Figure 11: Average values of selected Belgrade Groundwater Source parameters, by redox potential segment. (Legend: Eh – redox potential, d50 – aquifer grain size (50% of the grain-size distribution curve), v – entrance velocity, O2 – oxygen concentration in water, NO3 – concentration in water, Fe2+ – bivalent iron concentration in water, SO4 – concentration in water, and Ec – water conductivity).

 

Tab01

The Danube River

Kovin-Dubovac Drainage System

The area between Kovin and Dubovac occupies the southern part of the Banat Depression and comprises the alluvial plain of the Danube between these two towns, Fig. 12. Following construction of the Iron Gate 1 Hydroelectric Power Plant downstream from this area and the impoundment of the Danube in 1972, the groundwater levels in the riparian lands were permanently increased. This necessitated a protection system, which is comprised of dikes, drainage ditches, and a line of drainage wells. The role of the protection system is to maintain groundwater levels at pre-defined depths and prevent spreading of the Iron Gate 1 HPP reservoir impact further inland.

The lower part of the aquifer is made up of upper Quaternary/upper Pleistocene and Holocene gravels and sandy gravels, whose grain sizes vary to a large extent. They are overlain by medium-size gravels, sandy gravels and gravelly sands.

 

This sequence of sediments ends with semi-permeable silty sands, silts, and silty and riverine/bog clays. The average thickness of the aquifer strata is about 20 m, Fig. 13.

The size of d50 gravel sediment grains is generally from 0.5 to 20 mm, Fig. 14. The overlying sediments have a d50 grain size between 0.08 and 0.3 mm, and the semi-permeable aquifer roof sediments have a d50 grain size of 0.007 to 0.02 mm.

Water quality data were collected from 2010 to 2013. Table 2 shows these data on select wells of the drainage system shown in Fig. 12.

The average values of Eh were found to generally be in the range from 50 to 150 mV. Only a small number of samples measured more than 200 mV, Fig. 15. The low entrance velocities are a result of the fact that the wells are self-discharging.

 

fig12
Figure 12: Series of drainage wells along the left bank of the Danube between the towns of Kovin and Dubovac.

 

fig13
Figure 13: Typical section through the alluvial sediments between Kovin and Dubovac.

 

fig14
Figure 14: Grain-size distribution curves of the alluvial aquifer along the left bank of the Danube, between Kovin and Dubovac.

 

fig15
Figure 15: Average values of selected parameters of Kovin-Dubovac wells, by redox potential segment. (Legend: Eh – redox potential, d50 – aquifer grain size (50% of the grain-size distribution curve), v – entrance velocity, O2 – oxygen concentration in water, NO3 – concentration in water, Fe2+ – bivalent iron concentration in water, SO4 – concentration in water, and Ec – water conductivity).

 

Tab02

Knićanin-Čenta Drainage System

This drainage system is comprised of drainage ditches and self-discharging wells, built in the 1970's to protect riparian lands from elevated stages of the Danube, resulting from its impoundment for the purposes of the Iron Gate 1 HPP. It is located on the left bank of the Danube, near the mouth of the Tisa River, Fig. 16. The Town of Knićanin is protected by a series of pumped wells along the perimeter of the city towards the Tisa. Self-discharging wells are lined up towards the Danube. Their operation is based on stages that cause discharge into the drainage ditches, whose water levels are controlled by a pumping station. Water from the drainage system is pumped back into the Danube by a pumping station at Čenta.

The thickness of the alluvial aquifer is from 15 to 28 m. The grain sizes of the lower zone are coarser than those of the upper zone. The aquifer floor is comprised of virtually impermeable clays, while the roof is made up of silty and clayey semi-permeable sediments. The central part of the cross-section features occasional clay and silty semi-permeable lenses, Fig.17.

Medium-grain sands are dominant in the central and upper parts of the aquifer, while the lower part features sandy and fine-grain gravels, Fig. 18.

In general, the average values of Eh of the studied wells did not exceed 250 mV, Fig. 19; only one sample measured 273 mV, Table 3. Two wells at Knićanin exhibited elevated entrance velocities, as a result of pumping. The wells along the Danube and the Begej are self-discharging.

 

fig16
Figure 16:  Reach of the Danube between Knićanin and Čenta.

 

fig17
Figure 17: Section through alluvial sediments along the left bank of the Danube between Knićanin and Čenta.

 

fig18
Figure 18: Grain-size distribution curves of aquifer sediments between Knićanin and Čenta.

 

fig19
Figure 19: Average values of selected parameters Knićanin-Čenta wells, by redox potential segment. (Legend: Eh – redox potential, d50 – aquifer grain size (50% of the grain-size distribution curve),
v – entrance velocity, O2 – oxygen concentration in water, NO3 – concentration in water, as N, Fe2+ – bivalent iron concentration in water, SO4 – concentration in water, and Ec – water conductivity.

 

Tab03

 

Veliko Gradište Drainage System

The Veliko Gradište Drainage System is located within the Town of Veliko Gradište, along a road adjacent to a regulated riverbank, Fig. 20. The wells are equipped with pumps, which maintain pre-defined groundwater levels in the riparian lands. Only two wells have been sampled in a single campaign. The data were certainly insufficient, but were still used in the present research.

The aquifer is comprised of coarse-grain sand and gravel, Fig. 21: d50 from 1 to 11 mm, and d10 from 0.1 to 0.3 mm.

The single groundwater sampling campaign involving two wells revealed a relatively high redox potential, Eh, of some 320 mV. However, it needs to be corroborated by further analyses.

 

fig20
Figure 20:  Map of the Veliko Gradište site.

 

fig21
Figure 21: Schematic section through the Veliko Gradište Drainage System.

The Velika Morava River

Ključ Groundwater Source

The Ključ Groundwater Source provides water supply to the City of Požarevac. It is located along the Velika Morava River and comprises a portion of its alluvial plain, Fig. 22.

The aquifer is comprised of polycyclic riverine/lacustrine and alluvial sediments, whose total thickness is 15 to 20 m. The part of the aquifer from which groundwater is extracted is predominantly represented by sandy gravels. The thickness of these gravels in the Ključ area is from 6 to 11 m, and adjacent to the Velika Morava from od 8 to 12 m. The gravel sequence is locally overlain by sands, whose thickness is 2 to 6 m. The sandy sequence is covered by semi-permeable sediments, generally loess and clay, whose thickness ranges from 3 to 5 m. A typical lithological section through the sediments at the source is shown in Fig. 23, along with the line of the Velika Morava riverbed indicative of a good hydraulic contact between the groundwater and the river.

The above is corroborated by the grain-size distribution in the source area, Fig. 24. The aquifer roof and floor grains are up to a thousand times finer than the grains of the aquifer itself.

 

The groundwater at the Ključ Groundwater Source along the Velika Morava is characterized by a high redox potential, greater than 300 mV, along with a relatively high concentration of dissolved oxygen, Table 4, Fig. 25. There is virtually no bivalent iron, but nitrate concentrations were found to be elevated, suggesting pollution from the ground surface.

 

Comment on Oxic States

A summary presentation of the parameters of all the studied locations is shown in Fig. 26. The data were previously interpreted as representative by well, then the values averaged in the given Eh intervals (of 50 mV). The wells of the Belgrade Groundwater Source were dominant, where the prevailing redox potential was below 150 mV. There was a significant decrease in Fe2+ concentrations, and increase in d50 with increasing Eh.

Figure 27 shows the average values of the analyzed parameters for three most characteristic sites: Belgrade Groundwater Source (the Sava River), Kovin-Dubovac Drainage system (the Danube River), and Ključ Groundwater Source (the Velika Morava River).

 

fig22
Figure 22: Map of Ključ Groundwater Source.

 

fig23
Fig. 23:  Lithological section through the Ključ Groundwater Source.

 

fig24
Figure 24: Grain-size distribution of alluvial sediments in the Ključ Groundwater Source area.

 

fig25
Figure 25: Average values of selected parameters of the Ključ Groundwater Source near the City of Požarevac, by redox potential segment (Legend: Eh – redox potential, d50 – aquifer grain size (50% of the grain-size distribution curve), v – entrance velocity, O2 – oxygen concentration in water, NO3 – concentration in water, as N, Fe2+ – bivalent iron concentration in water, SO4 – concentration in water, and Ec – water conductivity.)

 

tab04

 

fig26
Figure 26: Average values of selected well parameters of all study areas, by redox potential segment. (Legend: Eh – redox potential, d50 – aquifer grain size (50% of the grain-size distribution curve), vprop  – recommended entrance velocity, O2 – oxygen concentration in water, NO3 – concentration in water, as N, Fe2+ – bivalent iron concentration in water, SO4 – concentration in water, and Ec – water conductivity).

 

fig27
Figure 27: Comparison of average values of several parameters of the groundwater and medium: sites on the Sava River (Belgrade), the Danube River (Kovin-Dubovac, Knićanin-Čenta), and the Velika Morava River (Požarevac).

LHR Variation: Case Study of the Belgrade Groundwater Source

A total of 18 new laterals were installed on five radial wells (hereafter: new wells) at the Belgrade Groundwater Source from 2005 to 2008. Their individual lengths were between 40 and 50 m. The pre-existing, "old" laterals were shut-off and decommissioned.

Since the very beginning, the new wells exhibited a capacity decline. A long-term study has been conducted to assess the rate of clogging of the new laterals, expressed via the increase in local hydraulic resistances. LHR variation monitoring, or the quantification of KLHR, indicated biochemical iron clogging of the five new wells.

Figure 28 shows the variation in LHR over time for the new wells. The relative temporal axis (horizontal axis) represents the period of operation of the new wells. The laterals of most of the wells were regenerated at least once during the study period (denoted by broken lines and R in the figure).

It is apparent that in the 6-7 years of operation of the new wells, LHR increased about 20 times, on average.

Initially, all wells exhibited an LHR of the same order of magnitude, which actually represented hydraulic resistances without the clogging effect. The resistance created due to clogging over time differed from well to well. Its magnitude was proportional to the groundwater Eh and Fe2+. The rate of clogging, KLHR, can readily be obtained from the plots in Fig. 28, via Eq. (9).

for09(9)

where: t1 is the selected point in time when the process is observed, and t0 is the time immediately after installation of laterals (or after regeneration). The estimated KLHR values of the five new wells are shown in Table 5.

 

Since 2005, the water utility has undertaken several regenerations of laterals on four of the five new wells. Based on available data, two intervals were selected and two KLHR calculations performed for well RB-5m, before and after regeneration. The somewhat slower post-regeneration increase in KLHR was attributable to the pre-existing microbial quasi-equilibrium resulting from prior operation of the well.

The assessment of the five new wells revealed an undoubted correlation between the rate of clogging and the chemical properties of the groundwater, primarily the iron concentration and redox potential. Figures 29 and 30 contain plots which indicate that the rate of clogging (KLHR) was proportional to the iron concentration in the groundwater, and inversely proportion to the redox potential.

It clearly follows from this example that iron clogging is of overriding importance for the well ageing process. Iron clogging dominated other types of clogging of radial well laterals. Based on assessments of the other wells at the same site, it is possible to generalize that in alluvial environments the rate of well ageing is accelerated in anaerobic and semi-anaerobic conditions (Eh < 150 mV). The analyses involved a large number of samples, ensuring a relatively high degree of confidence in the conclusions. However, the large areal extent of the Belgrade Groundwater Source (50 km of riverbank) and the diversity of geological and other conditions in its various parts resulted in data scatter, as apparent in the figures.

The rate of well clogging was found to be higher under conditions that corresponded to natural anoxic, rather than mildly aerobic, conditions. Estimation of the rate of well clogging is important for selecting the location of a future groundwater source and designing wells (screen type and material). The previously-discussed indicators also provide insight into the cost of maintenance.

 

fig28
Figure 28: LHR change after installation of new laterals, R – regenerated (Dimkić et al., 2011b, 2012 revised, supplemented).

 

tab05

 

fig29
Figure 29: KLHR as a function of iron concentration (new wells at the Belgrade Groundwater Source), Dimkić et  al., 2011b, revised.

 

fig30
Figure 30: KLHR as a function of redox potential (new wells at the Belgrade Groundwater Source),  Dimkić et  al., 2011b, revised.

Assessment of the Generalized Plot
of VCR = F(EH)

In engineering practice to date, great care has been taken to ensure the aquifer's filtration stability. In other words, entraining and removal of aquifer particles and their entry into the well need to be prevented. If not, normal operation of the well becomes threatened to the point of decommissioning. Filtration stability is ensured by limiting well screen capacity and entrance velocity. Allowable entrance velocities, which ensure the aquifer's filtration stability in the near-well region, have been studied by Istomina, 1957, Abramov, 1952, Cistin, 1965, Johnson, 1975, Gavrilko, Alekseev, 1985, Truelsen, 1988, Vuković, Soro, 1990, Vuković, Pušić, 1992, Dimkić, Pušić, 2008 and others.

However, the entrance velocities quantified in this manner do not ensure the absence of well ageing due to clogging, nor do they in the very least slow down this process. An LHR increase, manifested by an increased "parasitic" well drawdown, is an important consideration in well maintenance planning.

A large database is available at the Belgrade Groundwater Source on the history of well discharges, regenerations, numbers and lengths of radial well laterals, pumping tests, and groundwater redox potential and bivalent iron concentrations. The number of analyses varies from one to more than ten, depending on the well in question.

Forty-seven wells were selected for the present study: 42 "old wells" (whose laterals had not been replaced) and five new wells, whose laterals were replaced between 2005 and 2008 with stainless steel laterals, and where three or more analyses had been performed.

The focus was on finding out whether there was a correlation between Eh and LHR increase due to iron clogging. Local losses were determined and then their variation over time analyzed, with special emphasis on the variation (increase) in local drawdown (∆S) on an annual basis. The wells were categorized as shown in Table 6.

 

It was found that 15 wells exhibited a change of ∆S < 0.5 m/year and 35 wells a change of ∆S > 0.5 m/year. Average entrance velocities (v) were calculated for all the wells and plots constructed of entrance velocities as a function of redox potential, v = f(Eh), Fig. 31. The representation also included the allowable entrance velocity criterion, which is deemed to ensure filtration stability of the near-well region (several authors: Abramov, Sichardt, Kovacs, Gros...; Sichardt and Kovacs criteria were considered here, as they are the most stringent). Also depicted in the figure are the other study areas: Kovin-Dubovac, Knićanin-Čenta and Veliko Gradište along the Danube and the Ključ Groundwater Source adjacent to the Velika Morava, with a total of 22 wells.

The correlation between the groundwater redox potential, well entrance velocities and the rate of well ageing (or the rate of increase in local resistances) was used to determine the so-called critical, maximum allowable entrance velocities. In view of its nature, this concept is of both technical and economic nature. In the BGWS case, wells that exhibited an annual increase in local drawdown of ∆SAV = 0.35 m were deemed to be "good" wells. This was also the allowable level (i.e. a higher rate of well clogging should not be permitted) (Eq. 7).

Wells whose ∆S ≤ 0.5 m were used as benchmarks for the quantification of the critical entrance velocity, based on the adopted well ageing criterion. A table was generated, which also included select wells along the Velika Morava and the Danube. A total of 15 wells were assessed and used for this purpose. No re-calculation for ∆S = 0.35 m/year was made for the wells along the Velika Morava and the Danube. Instead, actual data were used as the wells along the Velika Morava exhibited virtually no losses, while those along the Danube measured extremely low discharges and could therefore have no substantial losses.

Given that the availability of data in the considered range was not uniform, the data were re-distributed as average values in selected intervals. The outcome of this interpretation is shown in Table 7.  The resulting points were plotted (Fig. 32) and connected by two lines: one for v (filtration stability), and the other for
v(∆S = 0.35 m/year.

 

Tab06

 

fig31
Figure 31: Well entrance velocities as a function of groundwater redox potential.

 

Tab07

Figure 32 is one of the outcomes of the study, which illustrates the correlation between Eh and well entrance velocities with regard to the considered alluvial aquifers in Serbia. The conclusion was that the velocities that ensure filtration stability of the near-well region do not meet the criterion of the allowable increase in local hydraulic resistances at low values of Eh (especially Eh < 200 mV). When Eh is greater than 300 mV, the two criteria become virtually equal.

In view of the fact that the focus was on well ageing caused by iron clogging, the correlation between Eh and Fe2+ was analyzed. Figure 33 shows averaged results for the studied wells along the three rivers. The correlation between these two parameters could not be doubted. For values of

Eh of 150 to 200 mV, the average Fe2+ concentration was of the order of 0.5 mg/l. If Eh was lower, the Fe2+ concentration increased rapidly.

Based on the established correlation, a plot was constructed of entrance velocities as a function of Fe2+ concentration, Fig. 34.

It was found that for the quantification of maximum allowable entrance velocities, the variability of local hydraulic resistances also needed to be taken into account. At nearly all the wells, this quantity was much smaller than if estimated using the formulas that ensure filtration stability of the near-well region. This was especially true of anoxic groundwater conditions, or in the present case for Eh ≤ 200 mV and Fe2+ ≥ 0.5 mg/L, Figs. 32 and 34.

The above shed some light on the phenomenon that has long puzzled specialists who studied well ageing at the Belgrade Groundwater Source. In well design, the entrance velocity criteria applied were those that ensured filtration stability of the near-well region and well screen. Here a clearer emphasis is placed on the impact of biochemical factors (above all the groundwater redox potential and iron concentration) on the intensity and rate of well clogging.

Data collected from the other study areas were added to the data from the Belgrade Groundwater Source, to provide a more complete picture of the effect of iron concentration and oxic state of groundwater on well ageing. The data presented above are related to study areas along three rivers (the Sava, the Danube and the Velika Morava) and they confirm the general prevalence of the same well clogging processes in alluvial environments.

For values of Eh greater than 200 mV and iron (Fe2+) concentrations less than 0.5 mg/L, the criteria derived from the filtration stability condition become dominant, while the effect of biochemical clogging is reduced and ultimately ceases.

 

Conclusions

Well ageing due to well-screen clogging is a long-known problem. The economic and technical significance of the clogging phenomenon is enormous. Mechanical clogging, due to excessive entrance velocities, has been studied extensively and the relevant criteria defined by old masters (Abramov, 1952, Cistin, 1965, Gavrilko and Alekseev, 1985, Johnson, 1975, Kovacs and Ujfaludi, 1983, Pietraru, 1982, Sichardt, 1928 and others).

 

 

Biochemical clogging (biofouling) occurs as a result of processes that take place in anoxic groundwater in the near-well region. A number of authors worldwide have considered the anoxic state and the processes that lead to well ageing. However, the objective of the long-term and broad-based research reported here was to establish a correlation between some of the most important biochemical parameters of groundwater (Eh, Fe) and mechanical parameters (groundwater entrance velocities, rate of clogging). As far as we are aware, this constitutes pioneering work on a global scale.

Research was conducted on several sites along three rivers in Serbia (the Sava River – Belgrade Groundwater Source; the Danube – riverbank sections from Kovin to Dubovac and from Knićanin to Čenta; and the Velika Morava – Ključ Groundwater Source near the City of Požarevac). All the tests revealed that there was virtually no biochemical clogging under oxic groundwater conditions (where the redox potential was greater than 250 mV).

In anoxic conditions (the redox potential range above 50 mV was studied), the main parameters that drive well clogging, apart from well entrance velocities, were found to include the groundwater redox potential and iron concentration.. The general expression is:

for10

where: v –entrance velocity, Fe – iron concentration in well water, Eh – groundwater redox potential, B – function of the rate of bacterial growth in the well, Γ – function of several parameters related to well design (with or without gravel pack, gravel pack characteristics, type and characteristics of screen slots), as well as the grain-size distribution of the aquifer.

Given the large number of parameters, different condition of the wells and a diversity of hydrogeological settings, it was impossible to establish very stringent correlations between the rate of clogging, and groundwater iron concentrations and redox potential. Still, the correlation outlined above is proposed. Allowable entrance velocities were also specified. Plots of the vcr = f(Eh, Fe) type are proposed as suitable and recommended for use in the design of both tube wells and radial wells. This is also deemed important for assessments of well regeneration and reconstruction methods.

The tests, both completed and ongoing, address a large number of other important aspects (nature of sediment formation, microbial activity of different micro-organisms, difference in redox potential between different parts of the porous medium, effect of the structural characteristics of well screens, well regeneration methods, etc.). The results of some of the tests will be reported shortly, while other tests are still to be conducted.

 

Acknowledgment

The present paper is an outcome of Project TR37014 "Methodology for the Assessment, Design and Maintenance of Groundwater Sources in Alluvial Environments Depending on the Aerobic State", which is funded by the Ministry of Education, Science and Technology Development of the Republic of Serbia.

 

 

fig32
Fig. 32: Well entrance velocities as a function of groundwater redox potential at a controlled annual increase in local drawdown of ∆S = 0.35 m/year.

 

fig33
Figure 33: Correlation between groundwater  Eh and Fe2+ in the alluviums of the Sava, the Danube and the Velika Morava (points represent average values by well).

 

fig34
Figure 34: Well entrance velocities as a function of Fe2+ concentration for a controlled annual increase in local drawdown of ∆S = 0.35 m/year.

References

Abramov, C.K. (1952). Method of Calculation and Selection of Filters for Drilled Wells (in Russian), Moscow

Barbič, F., Bracilović, D., Djindjić, Djorlijevski, S., Živković, J., Krajinčnić, B. (1974). Iron and manganese bacteria in Ranney wells. Water Research, Volume 8, Issue 11, November 1974, Pages 895-898

Cistin, J. (1965). Some Problems of Deformation of Mechanical Filters (in Slovakian). Vodohospod, Casopis 2

Cullimore, R., (1999). Microbiology of Well Biofouling. Lewis Publishers

Dimkić M. (2012): Processes in alluvial groundwater and their significance (in Serbian), 14th Serbian Symposium on Hydrogeology, 17-20 May 2012, Zlatibor, ISBN 978-86-7352-236-4, ppr. 5-10, 2012.

Dimkić, M., Pušić, M. (2008). Recommendations for Water Well Design Taking into Account Iron Clogging, Based on Experience Gained at the Belgrade Groundwater Source (in Serbian). Civil Engineering Calendar 2008, Vol. 40, pp. 430-496

Dimkić M., Taušanović V., Pušić M., Boreli-Zdravković Đ., Đurić D., Slimak T., Petković A., Obradović V., Babić R. (2007): Belgrade Groundwater Source, Condition and Possible Development Directions, IWA Publishing, Journal "Water Practice and Technology", ISSN 1751-231X, Volume 2, Issue 3, 2007.

Dimkić, M., Brauch, H.J., Kavannaugh, M. (2008). Groundwater Management in Large River Basins. IWA Publishing, London, UK

Dimkić, M., Pušić, M., Petković A., Boreli-Zdravković, Dj. (2010). Several natural indicators of radial well ageing at the Belgrade Groundwater Source, Part 1. Water Science and Technology, IWA Publishing,

Dimkić M., Pušić M., Majkić-Dursun B., Obradović V. (2011a): Certain Implications of Oxic Conditions in Alluvial Groundwater, Journal of Serbian Water Pollution Control Society "Water Research and Management", ISSN 2217-5237, Vol. 1, No. 2, p. 27-43, 2011.

Dimkić M., Pušić M., Obradović V., Djurić D. (2011b): Several natural indicators of radial well ageing at the Belgrade Groundwater Source, Part 2, Water Science & Technology, IWA Publishing, London, ISSN 0273-1223, vol. 63, no. 11, p. 2567-2574, 2011.

Dimkić M., Pušić M., Vidović D., Petković A., Boreli-Zdravković Đ. (2011c): Several natural indicators of radial well ageing at the Belgrade Groundwater Source, Part 1, Water Science & Technology, IWA Publishing, London, ISSN 0273-1223, vol. 63, no. 11, p. 2560-2566, 2011.

Dimkić M.., Pušić M., Obradović V., Kovačević S. (2012): The effect of certain biochemical factors on well clogging under suboxic and mildly anoxic conditions, Water Science & Technology, IWA Publishing, London, ISSN 0273-1223, vol. 65, no. 12, p. 2206-2212, 2012. (doi: 10.2166/wst.2012.129)

Dubinina, G. A. (1978): Mechanism of oxidation of divalent iron and manganese by iron bacteria growing in neutral medium, Mikrobiologiya 47, 59 1-599.

Gavrilko, V.M., Alekseev, V.S. (1985). Water Well Screen. Publishing House "NEDRA", Moscow, pp. 300-304

Houben, G. (2003) Iron Oxide Incrustation in Wells: Parts 1 and 2. Applied Geochemistry, Elsevier Science.

Houben, G., Treskatis CH. (2007). Water Well Rehabilitation and Reconstruction. McGraw Hill, ISBN-13: 978-0-07-148651-4, pp. 391

Istomina, V.S. (1957). Filtration Stability of Soils (in Russian). Gostroizdat, Moscow-Leningrad

Johnson, E.E. (1975). Ground Water and Wells: A Reference Book for the Water Well Industry. Edward E. Johnson, St. Paul, Minnesota

Jurgens, B.C., McMahon, P.B., Chapelle, F.H., Eberts, S.M. (2010): An Excel Workbook for Identifying Redox Processes in Ground Water. USGS. USA.

Kovacs, G.., Ujfaludi, L. (1983). Movement Of Fine Grains In The Vicinity Of Well Screens. Hydrological Sciences – Journal des Sciences Hydrologiques, 28, 2, 6/1983

Pietraru, V. (1982). Rapport general sur les crepines et filtres des puits. International Symposium on Hydraulic Problems of Ground Water Drainage, Gornji Milanovac

Petković A., Dimkić M., Boreli-Zdravković Đ. (2008): Measurement of groundwater redox potential: importance, concept and problems (in Serbian), 37th Conference on Current Water Use and Water Protection Issues "VODA 2008", 3-6 June 2008., Mataruška Banja, ISBN 978-86-904241-5-3, pp. 221-224, 2008.

Sichardt, W. (1928). Method of Stabilization of Drilled Wells (in German). Springer, Berlin

Truelsen, C. (1988), Groesste Brunnenleistung. Die Wassererschliessung: Grundlagen der Erkundung, Bewirtschaftung und Erschliessung von Grundwasservorkommen in Theorie und Praxis. 3rd ed., H. Schneider, ed., Vulkan - Verlag, Essen, West Germany

Vuković, M., Pušić, M. (1992). Soil Stability And Deformation Due To Seepage. Water Resources Publications, Littleton, Colorado, USA, p. 80

Vuković, M., Soro, A. (1990). Hydraulics of Water Wells, Water Resources Publications, Littleton, Colorado, USA, p. 354

www.Google Earth

 

 

River Sediment Transport in Serbia

Slobodan Petković1

 

1Jaroslav Černi Institute for the Development of Water Resources, Jaroslava Černog 80, Belgrade, Serbia

 

Abstract

River sediment transport within the territory of Serbia is discussed in the paper. The correlation between hydrological and sediment transport parameters was analyzed using data from a sediment database compiled for rivers where the relevant parameters are monitored and from which the sediment regimes have been determined for small, medium and large rivers. This facilitated an assessment of Serbia's approximate sediment budget.

Keywords: sediment transport, rivers, Serbian territory, sediment budget.

 

Introduction

 

Comprehensive studies of the production and river sediment transport in central Serbia are generally based on the essential link between erosion processes within drainage areas and sediment transport along the Serbian river network. The spatial scale of these processes is important; erosion processes tend to take place in mountainous regions, where transport processes begin via a network of torrential streams, and end in the largest lowland rivers. As a result, a river sediment transport assessment needs to encompass the entire river network in Serbia.

The territory of Serbia has a natural hydromorphological predisposition for the development of erosion processes, given that three quarters of the territory isare hilly or mountainous. There are numerous characteristic topographic features and steep slopes conducive to intensive erosion. The geological makeup is heterogeneous but with a large proportion of erodible rocks. On the other hand, the vegetation cover in hilly areas is not adequate to protect the soil from mass wasting. Consequently, there is an interaction of sorts between the three major drivers of erosion processes: energy, resistance and protection, which has resulted in the erosion patterns generally encountered in Serbia.

The goal of the present study of river sediment transport in Serbia was to determine the general sediment budget. Most of the sediment produced within Serbia, which becomes entrained along the river network, reaches the main watercourses that empty into two large rivers, which flow along the northern boundary of central Serbia. The two rivers are the Sava and the Danube. Their major tributaries include the Velika Morava, the Drina, the Kolubara, the Mlava, the Pek and the Timok. It should be noted that the Drina and the Timok define parts of Serbia's state border and thus receive sediment from neighboring countries—Bosnia and Herzegovina, and Bulgaria. Serbia's river network is shown in Fig. 1.

The land area of central Serbia is roughly 56,000 km2, and within it the river basin of the Velika Morava (Greater Morava), including the drainage areas of the Južna Morava (South Morava) and Zapadna Morava (West Morava), at whose confluence the Velika Morava originates, is about 38,000 km2. It is for that reason that the Velika Morava River Basin plays a major role in river sediment transport studies and in the determination of the general sediment budget. In this regard, the fact that most of the gauging stations that monitor sediment are located within the Velika Morava River Basin is a favorable circumstance, which facilitates river sediment transport and sediment budget assessments.

Sediment Database

The periods for monitoring hydrological parameters in Serbia vary, as a result of historical circumstances. The gauging stations on the largest river, the Danube, feature the longest time series, given that observations of that river began in the early 20th century. Monitoring of other rivers started much later, after World War II. However, there are no significant differences in sediment monitoring, given that nearly all the sediment data were collected after the year 1945. In central Serbia, the longest time series are provided for the Velika Morava, the Južna Morava and the Zapadna Morava Rivers. Over time, the monitoring network of the National Hydrometeorological Service of Serbia (RHMZ) was expanded to include more rivers. Unfortunately, sediment monitoring has not been continuous at all stations, so the time series are not chronologically the same. Additionally, sediment monitoring data are almost exclusively related to suspended sediment, except for short-term bedload measurements on the Velika Morava and the Drina.

The suspended-sediment monitoring method applied is comprised of suspension sampling from the main stream and laboratory analysis of the samples, to determine the concentration of the solid phase in the water. In Serbia, samples are generally collected once a day (in the morning), and then taken to the lab. It should be noted that the suspended-sediment monitoring procedure followed at all RHMZ gauging stations has not been adequate for quite some time. Namely, suspension samples are always collected from the same point in the hydrometric cross-section, in an insufficiently standardized manner—using one-liter bottles.

Also, the samples are filtered and measurements made in makeshift conditions, which limit accuracy.

River sediment sampling once a day, at a particular time, cannot provide realistic and representative information about suspended-sediment concentrations and transport, especially in the case of small rivers. It is well-known that the largest amounts of sediment are conveyed by flood waves and that in most cases flood wave peaks tend to occur between the scheduled sampling times. As a result, daily sediment concentrations are recorded, which are much lower than the maximum values during the day (e.g. at the time of a flood wave peak). The ultimate consequence of this inadequate suspended-sediment monitoring approach is an unrealistically low level of the summary annual rate of sediment transport along the considered river.

When suspended-sediment monitoring data collected in the above manner are analyzed, the size of the river and the hydrological regime also need to be considered. Hydrological and sediment-related changes in large alluvial rivers tend to be relatively slow, such that same-time monitoring once a day will generally provide representative data. However, the situation is very different in the case of small rivers, especially those whose hydrological regime is torrential in nature. It is often the case that recorded daily sediment concentrations are much lower than the maximum values at the time of flood wave peaks. As such, the results need to be corrected on the basis of available general knowledge of the hydrological and sediment regimes of the river in question, as well as through analogy with other rivers of similar characteristics.

Fig01
Figure 1: Serbian river network.
 

Review of Sediment Transport Data on Some Rivers in Serbia

Given the extensive suspended-sediment monitoring data on Serbian rivers, this paper addresses only some watercourses, being specific from a sediment transport perspective and representative of different river categories: the Velika Morava is a large river, the Lim is medium-sized, and the Vlasina and the Toplica (tributaries of the Južna Morava) represent rivers with a catchment area of up to 1000 km2.

 

Velika Morava River

The Velika Morava is the largest river in central Serbia, having a catchment area of some 37,000 km2.

This river is sometimes referred to as "the muddy river", because of considerable amounts of suspended sediment resulting from the substantial erosion potential of its drainage area. There are several distinct erosion-prone areas in the Morava River Basin, particularly in the catchment area of the Južna Morava. On the other hand, the suspended-sediment transport capacity of the Velika Morava is considerable, such that most of the sediment delivered by its tributaries reaches the confluence of the Velika Morava and the Danube.

As in the case of all natural watercourses, the suspended-sediment transport regime of the Velika Morava varies to a large extent during the year. Since sediment transport depends, above all, on the hydrological conditions of the river, there is a certain general (but not strong) correlation between river discharge and suspended sediment transport.

Suspended sediment concentrations (C) in the Velika Morava span a broad range, between 10-2 kg/m3 and 10 kg/m3. In other words, the ratio of maximum to minimum concentrations of suspended sediment is of the order of 103. However, the most frequent suspended sediment concentrations in the Velika Morava are 0.01 kg/m3.

Figure 2 gives the average annual river discharge (Qav) and the annual rate of suspended sediment transport (Ps) at the Velika Morava River, between 1961 and 2003. The values of Qav range from 100 to 350 m3/s, while those of Ps are from 0.40 to 13.00 million tons/year. The chronological chart exhibits a distinct downward Ps trend during the study period (1961-2003). At the beginning of that period (from 1961 to 1968), the values were Ps ≥ 5 million tons, from 1971 to 1981 2.5 < Ps < 4.5 million tons, and in recent times (1981-2003) 0.5 < Ps < 3.8 million tons. Several hypotheses can expain this significant decline in the Ps trend:

Initially (before the year 1970), the sediment measurement approach was not sufficiently accurate or reliable, resulting in higher-than-real suspended sediment concentrations. Consequently, only the period from 1971 to 2004 can be deemed representative for sediment transport assessments of the Velika Morava.

Erosion control and hydraulic engineering projects in the Velika Morava River Basin have reduced sediment production and retained a significant proportion of the sediment within the drainage areas of tributaries, thus decreasing the amounts of sediment delivered to the Velika Morava.

 

Fig02
Figure 2: Chronological bar chart of annual mean river discharge (Qav) and annual suspended sediment transport (Ps) of the Velika Morava, Ljubičevski most, 1961-2003.

In order to come up with realistic and representative annual rates of suspended-sediment discharge by the Velika Morava, the hydrological and sediment transport parameters during the period from 1960 to 2003 needed to be assessed in parallel. This comparative analysis led to the following conclusions:

  • The values of Ps > 5 million tons tend to occur only in extreme situations (years with one or more major flood events).
  • The values of Ps < 2 million tons tend to occur in very dry years.
  • Based on the above, the level of annual sediment discharge of the Velika Morava is in the range from 2 to 5 million tons. In average hydrological conditions, the representative rate is Ps = 4 million tons/year.

The Velika Morava is a rare river in Serbia where bed-load has been assessed. The bed-load transport assessment was based on monitoring data and hydraulic analysis of the sediment transport capacity of the river. The assessment included the determination of all bed-load transport parameters, encompassing hydrological, hydraulic, morphological and sediment transport aspects. Bed-load monitoring has been conducted under low to medium-high flow conditions. Unfortunately, only one measurement was made under high flow conditions and that fact was a data interpretation constraint. Based on available data, the correlation between sediment transport and river discharge Gv(Q) was determined for two localities: the first is beyond the zone of influence of the Danube impoundment (at Ljubičevski Most), and the second in the zone of the mouth of the Velika Morava (Fig. 3).

Figure 4 shows bed-load transport duration curves for these two localities, based on the adopted function Gv(Q) and the water discharge duration curve at the gauging station of Ljubičevski Most.

 

Integration of the bed-load transport duration curves resulted in the following annual rates of bed-load transport:

  • Ljubičevski Most 210,000 tons/year
  • River mouth 19,000 tons/year.

The conclusion was that 210,000 tons/per year was a realistic indication of the annual rate of bed-load transport upstream from Ljubičevski Most. On the other hand, due to the effect of the impoundment, only 10% of the Velika Morava bed-load was deemed capable of reaching the river mouth to the Danube.

 

Lim River

The Lim River is a tributary of the Drina, coming into Serbia from Montenegro. Suspended sediment in the Lim was measured between 1963 and 2002 at Prijepolje Station, where the upstream catchment area is 3160 km2. Due to its hydrological regime, the highest suspended sediment concentrations (С) were registered from February to April, and the lowest from September to November. The extreme daily values of sediment concentration were: minimum of the order of 10-3 g/l, and maximum 9.3 g/l.

Variations of river discharge and sediment concentration in 1986, which was characterized by high sediment loads, are presented on Figures 5 (entire year) and 6 (flood wave in February 1986, Cmax= 9.2 g/l).

Based on the available suspended-sediment database, taking into account the sediment transported by flood waves (one of which is depicted in Fig. 6), a correlation was established between the sediment concentration (C) and river discharge (Q). The values of C and Q were used to assess the magnitude of sediment transport (P = C x Q), given in Table 1.

The suspended sediment duration curve (Fig. 8) was determined from the function P(Q) and  the river discharge duration curve (Fig. 7).

 

Fig03
Figure 3: Sediment transport (Gv) as a function of the Velika Morava discharge (Q), at Ljubičevski Most and the river mouth.

 

Fig04
Figure 4: Bed-load transport duration curves, at Ljubičevski Most and the river mouth.

 

Tab01

Fig05
Figure 5: 1986 annual hydrograph and sediment load graph, Lim River at Prijepolje.

 

Fig06
Figure 6: Hydrograph and sediment load graph of flash-flood wave in February 1986, Lim River at Prijepolje.

 

Fig07
Figure 7: Water discharge duration curve, Lim River at Prijepolje.

 

Fig08
Figure 8: Sediment transport duration curve, Lim River at Prijepolje.

 

Integration of the function P(t) resulted in the annual suspended sediment discharge on the Lim River: P = 1,123,000 t/year.

In the case of the Lim River it was possible to check whether the estimated annual rate of sediment transport was realistic. Namely, the Potpeć Reservoir retains most of the sediment load of the Lim River. The reservoir sedimentation was surveyed a few times between 1967 and 1999 (Table 2).

 

Tab02

 

Table 2 shows that the Potpeć Reservoir had a serious sedimentation problem, given that more than 40% of its storage volume was lost between 1967 and 1999. On the other hand, it is also evident that the sedimentation rate has slowed down over the lifetime of the reservoir. Initially, the sedimentation rate was 0.9 million m3/year, but subsequently declined to 0.35 million m3/year.

The reservoir sedimentation data could also be assessed in the context of sediment discharge estimation. If the annual suspended sediment discharge is enlarged by 10% (usual proportion of bed-load in total sediment load), the total sediment load is 1,200,000 tons/year or 1,000,000 m3 (density of sediment deposits is 1.2 t/m3). The comparison of these figures with the average annual sedimentation rate of the Potpeć Reservoir in the initial period (890,000 m3) demonstrated that the estimates were realistic, keeping in mind that the reservoir could not retain more than 80% the total sediment load (or some 800,000 m3) since a part went through the power plant turbines and over the dam.

 

Vlasina and Toplica rivers, tributaries of the Južna Morava

The Vlasina River drains an area of 880 km2. The suspended sediment database covers the period from 1962 to 2002, and reveals that the highest suspended sediment concentrations (C) were recorded in the spring, from March to May, and the lowest from September to November. The extreme daily values of C were: minimum of the order of 10-4 g/l, and maximum 101 g/l, which is a broad range.

Variations of river discharge and sediment concentration in 1976, which was characterized by high suspended sediment loads, are presented in Figures 9 (entire year) and 10 (flood wave in June 1976, Cmax= 23 g/l).

The Toplica River, up to the gauging station Pepeljevac, where sediment is monitored, drains an area of some 1000 km2. The suspended sediment database covers the period from 1973 to 2002. The highest sediment concentrations (C) were recorded from February to May, and the lowest from September to November. The range of values is similar to those of the Vlasina River.

Figure 11 presents the variations in river discharge and suspended sediment concentration in 1979, during which very high suspended sediment loads were recorded. Similar data are presented on Figure 12 for a flood wave in November 1979 (Cmax = 35 g/l).

The total annual suspended-sediment loads transported by the Vlasina and the Toplica were determined from sediment transport duration curves P(t), prepared from the P(Q) relations (shown on Figs. 13 and 14) and taking into account the duration of river discharges (given for both rivers on Fig. 15).

River discharge duration curves of the Vlasina and the Toplica (Figure 15) are very similar, due to their comparable size and geomorphology. However, the Vlasina is of a more torrential nature, with a wider span between low and high flows (e.g. major flood in 1988 had peak discharge as high as 500 m3/s or more). The Toplica features a broader range of low and medium flows, but a narrower range of high flows.

 

Fig09
Figure 9: Variation in river discharge and suspended sediment concentration in 1976, Vlasina River at Vlasotince.

 Fig10
Figure 10: Hydrograph and sediment load graph for a torrential flood wave in June/July 1976, Vlasina River at Vlasotince.

 

Fig11
Figure 11: 1979 annual hydrograph and sediment load graph, Toplica River at Pepeljevac.

 

It is well-known that the river sediment regime depends on both production of eroded sediment in the catchment area and the hydrological regime of the river. The functions S(Q) of the Vlasina and the Toplica (Figs. 13 and 14) show that the Vlasina exhibited higher levels of S (even above 10 kg/s), especially in the high-flow range. On the other hand, the torrential nature of this river resulted in higher water discharge and sediment transport rates at high flows. This explained the difference between the sediment transport duration curves of these two rivers, as shown in Fig. 16.

 

Fig12
Figure 12: Hydrograph and sediment load graph of a torrential flood wave in November 1979, Toplica River at Pepeljevac.

 

Fig13
Figure 13: Suspended sediment transport as a function of river discharge, Vlasina River at Vlasotince.

 

Fig14
Figure 14: Suspended sediment transport as a function of river discharge, Toplica River at Pepeljevac.

 

Fig15
Figure 15: River discharge duration curves of the Vlasina River and the Toplica River.

 

Fig16
Figure 16: Sediment transport duration curves of the Vlasina River and the Toplica River.

Overview of Suspended-sediment Transport Along Small and Large Watercourses

Long-term suspended sediment monitoring has been conducted on ten rivers in Serbia, such that there is a relevant sediment database. The monitoring network covers a broad range of catchment sizes, from 145 to 37,320 km2. This has facilitated assessments of the effect of catchment size on sediment transport processes and provided a general picture of the processes.

Figure 17 shows the correlations between unit sediment transport (g) and catchment size (Аb) for all the studied rivers. The g(Аb) envelopes for Serbia, determined on the basis of the sediment transport and reservoir aggradation data, are also indicated. It is obvious that most of the studied rivers are in the middle of the envelopes of extreme values of g(Аb). It is also evident that the span between two envelopes includes all the (g, Аb) pairs, meaning that the span reflects the characteristics of sediment formation and transport within the territory of Serbia.

 

Approximated Sediment Budget for Serbian Territory

Serbia's sediment monitoring network is not all-encompassing, such that not all the rivers which are significant contributors to the sediment budget are covered. However, it was possible to approximate Serbia's sediment budget based on the existing sediment database, along with the erosion map and the hydrological characteristics of major watercourses.

The largest watercourses in Serbia are transboundary rivers: the Danube, the Sava and the Tisa. Extensive databases are available, mostly relating to the period after the Danube was impounded and a dam built for the Iron Gate Hydroelectric Power Plant (one of the largest in Europe).

The sediment budget along the Danube through Serbia was determined from suspended sediment data on the Danube, the Sava and the Tisa, taking into account the morphological changes within the impoundment. As it enters Serbia, the Danube annually delivers 6.7 million tons of sediment on average. The Sava conveys 3.0 million tons and the Tisa 4.4 million tons. Downstream from Belgrade, after the mouth of the Velika Morava, the Danube discharges 18.5 million tons of suspended sediment, most of which remains within the Iron Gate Reservoir. Only about 3 million tons of suspended sediment passes the dam.

Apart from the transboundary rivers (the Danube, Sava and Tisa), the largest domestic river is the Velika Morava. The available sediment database on its river basin allowed an assessment of the summary suspended-sediment budget. The annual averages, related to the long - term period, were found to be:

  • Južna Morava 2,500,000 tons
  • Zapadna Morava 1,500,000 tons
  • Velika Morava 4,000,000 tons

It became apparent that the Južna Morava transports more sediment than the Zapadna Morava, even though the difference between the sizes of the two drainage areas is only 4% in favor of the Južna Morava. This means that much more eroded sediment is produced in the drainage area of the Južna Morava, compared to the Zapadna Morava. It should also be kept in mind that suspended sediment is conveyed down the river, such that there is a certain continuity to sediment transport. Of course, that does not mean that the sediment does not become deposited along the way (in places where the transport capacity is low), or that there are no local increases (due to fluvial erosion), but such occurrences have no major impact on the annual sediment budget.

Fig17
Figure 17: Unit sediment transport as a function of catchment size.

 

Another major watercourse within Serbia's river network is the Drina, which forms a large part of the border between Serbia and Bosnia and Herzegovina. The sediment database for this river is not sufficiently representative, but in its case another fact is of particular importance, as this watercourse is a major contributor to the approximated sediment budget. Namely, there are many impoundments in the Drina River Basin, where sedimentation is monitored and allows the continuity of sediment transport to be assessed. As a result, it was determined that the average annual rate of sediment transport in the Drina River Basin declined considerably following the formation of a system of reservoirs (from 2.5 million tons in the middle course to 0.5 million tons in the lower course).

For smaller rivers within the territory of Serbia, apart from existing sediment data, information on the production of eroded sediment and the hydrological characteristics of the drainage areas were also used.

Serbia's river sediment budget was approximated on the basis of the existing sediment database, taking into account the erosion map and hydrological characteristics of significant watercourses, as interpreted in Fig. 18.

 

Fig18
Figure 18: Sediment budget of Serbia’s river network.

Conclusions

The territory of Serbia and its natural geomorphology are conducive to erosion processes, given that three quarters of the territory is hilly or mountainous. Consequently, most domestic rivers feature significant sediment discharges. However, it should be noted that the largest watercourses in Serbia are transboundary rivers—the Danube, the Sava and the Tisa. Most of the sediment transport takes place along these rivers.

The assessment presented above shows the sediment regime of several characteristic watercourses, including large, medium and small rivers. In the large river category, sediment transport along the Velika Morava, the largest domestic river, is discussed. The conclusion is that sediment transport has recently been reduced. Erosion control and hydraulic engineering projects in the Velika Morava River Basin have decreased the production of eroded sediment and sediment is trapped within impoundments. The impoundments also retain a considerable amount of sediment delivered from the drainage areas of tributaries.

The Lim River was used as an example of a medium-sized river. The case study of that river is interesting because there is an impoundment where aggradation is monitored. A comparison of sediment measurement data and aggradation data revealed a relatively good match. Consequently, the conclusion is that Serbia's sediment database constitutes a sound basis for all river management activities.

The river sediment regimes of the Vlasina and the Toplica were assessed in the small river category. An especially interesting conclusion of that case study is that the hydrological and sediment regimes are in agreement. A more pronounced torrential nature of the Vlasina is reflected in higher rates of sediment discharge, especially in the high-flow range.

Serbia's river sediment monitoring network is not all-encompassing, such that not all the watercourses which are significant contributors to the sediment budget are covered. However, based on the existing sediment database, taking into account the erosion map and hydrological characteristics of significant watercourses, it is still possible to assess Serbia's approximated sediment budget. In that regard, the role of large river reservoirs in the sediment budget needs to be highlighted. There are several tens of such reservoirs in Serbia.

References

Babic Mladenovic M., Kolarov V. and V. Damjanovic (2013): Sediment regime of the Danube River in Serbia, International Journal of Sediment Research, ISSN 1001-6279, Vol. 28, No. 4, pp. 470-485, 2013.

Petković S. (1995): Origins of sediment transport in the Južna Morava River Basin (in Serbian). Monograph. University of Belgrade, Faculty of Forestry.

Petković S. (1996): Origins of sediment transport in the Zapadna Morava River Basin (in Serbian). Monograph. University of Belgrade, Faculty of Forestry.

Petković S., Dragović N., Marković S. (1999): Erosion and sedimentation problems in Serbia. Hidrological Sciences-Journal-des Sciences Hydrologiques, 44 (1), 63-77

Petković S. (2001): Contemporary approaches to river engineering based on harmonized water resources management and environmental objectives (in Serbian). Vodoprivreda, Nos. 189 -194. Belgrade.

Petković S. and M. Babić-Mladenović (2006): Origin and transport of bed-load along the Južna Morava River (in Serbian). Collection of papers of the Serbian Hydraulic Research Association, Belgrade.

Petković S. (2006): Importance of erosion control and flash-flood protection in water resources management (in Serbian). Water and Sanitary Technology, Issue 3. Belgrade.

Petković S. (2007): River sediment yield and sediment budget in the river network of Serbia. Proceedings of the 10th International Symposium on River Sedimentation, Moscow.

 

 

Diversity of Aquatic Macroinvertebrates in Streams in the Belgrade Region (Does Different Stream Types Matter?)

Vanja Marković1, Ana Atanacković1, Katarina Zorić1, Marija Ilić1, Margareta Kračun-Kolarević1, Bojana Tubić1

 

 

 

1 University of Belgrade, Institute for Biological Research “Siniša Stanković”, Despota Stefana Blvd. 142, 11060 Belgrade, Serbia; E-mail: This e-mail address is being protected from spambots. You need JavaScript enabled to view it

 

 

Abstract

The paper presents the study of macroinvertebrate diversity in the Belgrade region. The study is based on material collected in September of 2012. A total of 21 localities, belonging to 5 different stream types are analyzed. A total of 65 taxa were identified. Oligochaeta were found to be the most diverse group overall, as well as in each analyzed watercourse type. Besides Nematoda and Chironomidae, tubificid worms Limnodrilus hoffmeisteri Claparede, 1862 and Potamothrix hammoniensis (Michaelsen, 1901), as well as snail Physella acuta Draparnaud, 1805 were common for all analyzed stream types. Although a low number of commom taxa in analyzed types could suggest faunistical differences, the performed multivariate analysis (DCA) could not separate watercourses in the Belgrade region in respect to their macroinvertebrate fauna. Such faunistical structures with no difference in respect to water types could be due to deteriorated habitats in this urban area. For more reliable analysis, prolonged and more detailed sampling is needed.

Keywords: macroinvertebrates, streams, tipology, urban area, Belgrade.

 

 

Introduction

 

Urbanization is a process which alters the physical and chemical characteristics of streams as well as causing significant biological and ecological degradation (Cuffney et al., 2010). Such deterioration of stream habitats and consequently its biota, caused by urbanization nowadays is recognized as „urban stream syndrome"(Walsh et al. 2005). Accordingly, an urbanized and densely populated area such as the Belgrade region, inevitably contributes to a wide range of anthropogenic impacts to its aquatic habitats. As the most important, hydromorphological pressures (regulation and channelization, bank reinforcement and embankments, sediment/sand extraction), organic and nutrient pollution (communal and urban wastewaters, agricultural drainages) and industrial and toxic pollution (industrial wastewaters, medical waste), should be noted. Besides, sometimes overlooked sources of toxicants are heavily trafficked roads, which some authors consider as one of principal pollution sources in many urban areas (Beasley and Kneale, 2002). Despite expansion in the field of urban ecology and hydroecology (Vermonden, 2010; Pickett et al., 2011), in Serbia there is a general lack of data even for its largest urban area – the Belgrade region.

Belgrade is located at the confluence of two large European rivers – the Danube and the Sava. In addition, there is also a number of various smaller watercourses. While large rivers – the Danube and the Sava - are well researched, as a result of research conducted in the urban region (for example Kalafatić et al., 1997; Martinović-Vitanović et al., 1999; Jakovčev-Todorović et al., 2005; Popović et al., 2013), or as a part of broader research (for example Paunović et al., 2007; Paunović et al. 2008), whereas the smaller streams have been neglected. With scarce data, although it is proved that it could be valuable habitats for aquatic macroinvertebrates (Vermonden, 2010). Marković et al (2013) provided an overview of freshwater snail fauna in Belgrade Region. The ecological status of smaller Belgrade watercourses was assessed by Kračun et al. (2013), while Dragićević et al. (2010) estimated the water and habitat quality of the Topčider River.

Our aim in this paper is to contribute to the knowledge of aquatic macroinvertebrate fauna of this urban region, and to test if the differences in its diversity could be due to its different water types.

Material and Methods

In September 2012, during the low water period, as a part of regular surface water quality monitoring in the Belgrade region, conducted by the Belgrade Institute of Public Health and the Institute for Biological Research "Siniša Stanković", benthic macroinvertebrate samples were taken. Semi-quantitative sampling was done by using a hand net (25x25 cm, 500 µm mesh size) and a Van Veen grab (270 cm²). Where possible, a multi-habitat sampling procedure (AQEM, 2002; Hering et al., 2004) was applied. The samples were preserved by using 4% formaldehyde solution and further processed in the laboratory. Identification was done to species-level for the majority of taxa using appropriate taxonomic keys.

The Belgrade region, with 1.7 million residents living in a 3223 km² metropolitan area (530 inhabitants per km²), is the largest and the most populated urban area in Serbia. Besides the Danube and the Sava Rivers, many smaller rivers, streams and canals and the left and right tributaries of these large rivers contribute to a very unevenly developed hydrographic network. Furthermore, geomorphology with its two main regions – the southern hilly Šumadija/Balkan region and the northern flat Pannonian Plain region, contributes to the unique hydrographic network of this urban area.

Five different water types according to current regulations/legislative (Official Gazette, 74/2011) were investigated. A list of sampling sites and watercourses with corresponding water types is provided in Table 1. Type 1 watercourses are large lowland rivers with a domination of fine substrate; Type 2 are large rivers with a medium-sized substrate, apart from the Pannonian rivers;

Type 3 are small to medium watercourses with a domination of larger substrate fraction, in altitudes below 500 m.a.s.l.; Type 6 are either watercourses which are not covered by current legislative (Official Gazette, 96/2010), or small watercourses outside the Pannonian Plain, apart from Type 3 and Type 4; Type 8 includes artificial water bodies – canals.

Generally, all watercourses in this region are under intense anthropogenic pressures. The canals in the northern, Pannonian region are exposed to heavy organic\nutrient pollution, mainly from agricultural land drainage, but also from households and slaughterhouses in this part of the region. Streams in the southern, Šumadija region, particularly those in the metropolitan area are under more intense urban pressure, with a higher proportion of impervious surfaces (due to "urban wash-off" which commonly occurs; Duda et al., 1982), in addition to communal, industrial and toxic pollution. Large rivers such as the Sava and the Danube, into which all the other smaller streams in the region flow, act as the main overall collectors in addition to being exposed to all the above mentioned pollutants.

It should be noted that, excluding the Danube and the Sava, the majority of the investigated watercourses are situated in the southern, Šumadija region. In the northern, Pannonian region, except for the Obrenovac Canal, all the investigated watercourses belong to Type 8 (canals).

Detrended correspondence Analysis (DCA; Hill and Gauch, 1980) was performed on a 21-by-65 samples-by-taxa (presence/absence) data matrix. The obtained ordination biplot, consisting of points representing taxa and squares representing samples, has exhibited their multivariate relations. The calculation was done using FLORA software (Karadžić et al., 1998; Karadžić, 2013).

Tab01

Results and Discussion

A total of 65 taxa were identified in this investigation (Table 2). Oligochaeta were found to be the most diverse group with 17 taxa, followed by Crustacea and Gastropoda with 10 and 9 taxa, respectively. The diversity of other main macroinvertebrate groups was significantly lower (Table 3) These were found to be Oligochaeta and Crustacea, with 7 taxa each (Table 3). The Oligochaeta were the most diverse group in all analyzed watercourse types. In the large lowland rivers with a domination of fine substrate (Type 1) the most diverse group, besides worms, was Crustacea, while in all other types it was Gastropoda (Table 3). Regarding localities, the greatest diversity (and the only one with >15 taxa) was present at Stepojevac (Beljanica) with 21 taxa. On the other hand at the Dobanovci hunting area locality (Galovica) only one taxon was found. Low diversities (<5 taxa) were recorded in samples from Sopot Stream and the Obrenovac Canal (2 taxa, each), as well as the PKB Canal and the Barička Stream (3 taxa each).

Regarding the different stream types, overall diversity ranged from 9 taxa (Type 6) to 34 taxa (Type 8) (Figure 1).

 

Tab02

 

Tab03

 

Fig01
Figure 1: Total number of taxa found in different stream types.

In total, only five taxa were present in all the stream types. Besides Nematoda and Chironomidae, tubificid worms Limnodrilus hoffmeisteri and Potamothrix hammoniensis, and snail Physella acuta were common for analyzed stream types (Table 2). Tubificids L. claparedeanus Ratzel 1868 and L. udekemianus Claparede, 1862 as well as biting midges (Ceratopogonidae), were present in all stream types, except in Type 6, i.e. the type with the lowest overall diversity (Table 2).

A relatively low number of common taxa among the analyzed types could suggest noticeable faunistical differences. However, the performed multivariate analysis (DCA) does not provide support for the above mentioned, as the obtained biplot reveals overlapping of samples belonging to different stream types (Figure 2). However, some differences, regarding localities could be noted along the first DCA axys, particularly between the Danube Batajnica locality (1) on the left side, and the Kotež -Vizelj locality (19) on the right. The separation arises primarily due to the presence of taxa Psammoryctides barbatus (Grube, 1861) (code Psa bar), Isochaetides michaelseni Lastockin, 1936) (Iso mic) and Corophium curvispinum (Sars,1895) (Cor cur) in the first locality, i.e. due to the presence of Eiseniella tetraedra (Savigny, 1826) (Eis tet), Dugesia lugubris Schmidt, 1861 (Dug), Alboglossiphonia heteroclita (Linnaeus, 1758) (Alb), Bythinia tentaculata (Linnaeus, 1758) (Byt), Sympetrum sp. (Sym) and Caenis horaria (Linnaeus, 1758) (Cae hor) in the second (Vizelj-Kotež).

The performed multivariate analysis could not separate watercourses in the Belgrade Region, in regards to the corresponding types. As the main reason behind the revealed uniformity could be urbanization and other anthropogenic activities, resulting in a reduction in the number of less tolerant taxa, together with an increase in the diversity and significance of pollution-tolerant taxa (Walsh et al., 2005; Smith and Lamp, 2008). A faunistic shift is noticeable in the overall diversity of the main taxonomic groups, with Oligochaeta (ie tubificids) as the most diverse members of the macroinvertebrate community in all the analyzed types of watercourses. The performed ecological assessment analysis (Kračun et al., 2013) showed poor overall ecological status/potential of these urban watercourses, indicating deteriorated habitats and communities.

However, it should be pointed out that the number of analyzed samples/watercourses is limited, especially in the case of types 2 and 6 (only two each), which could be the reason for its lower general diversity, compared to the other types. For more reliable analysis, prolonged and more detailed sampling is needed.

 

Acknowledgments

This study was supported by the Institute of Public Health Belgrade.

 

Fig02

Figure 2: DCA performed on the 21 sample x 65 taxa data matrix (presence/absence); localities and taxa codes are provided in Tables 1 and 2, respectively.

References

AQEM Consortium, (2002). Manual for the application of the AQEM system. A comprehensive method to assess European streams using benthic macroinvertebrates developed for the purpose of the Water Framework Directive. Version 1.0 (www.aqem.de), February 2002, 202 pp.

Beasley, G. and Kneale, P. (2002). Reviewing the impact of metals and PAHs on macroinvertebrates in urban watercourses. Progress in Physical Geography, 26 (2): 236-270.

Cuffney, T. F., Brightbill, R. A., May, J. T. and Waite, I. R. (2010). Responses of benthic macroinvertebrates to environmental changes associated with urbanization in nine metropolitan areas. Ecological Applications, 20 (5): 1384-1401.

Dragićević, S., Nenadović, S., Jovanović, B., Milanović, M., Novković, I., Pavić, D. and Lješević, M. (2010). Degradation of Topciderska river water quality (Belgrade). Carpathian Journal of Earth and Environmental Sciences, 5 (2): 177-184.

Duda, A. M., Lenat, D. R. and Penrose, D. L. (1982). Water Quality in Urban Streams: What We Can Expect. Journal (Water Pollution Control Federation): 1139-1147.

Hering, D., Verdonschot, P.F.M., Moog, O. and Sandin, L. (eds) (2004). Overview and application of the AQEM assessment system. Hydrobiologia 516: 1–20.

Hill, M.O. and Gauch, H.G. (1980). Detrended Correspondence Analysis: An Improved Ordination Technique. Vegetatio 42: 47–58.

Jakovčev-Todorović, D., Paunović, M. M., Stojanović, B., Simić, V. M., Đikanović, V. and Veljković, A. (2005). Observation of the quality of Danube water in the Belgrade region based on benthic animals during periods of high and low water conditions in 2002. Archives of Biological Sciences, 57 (3): 237-242.

Kalafatić, V., Martinović-Vitanović, V. and Tanasković, M. (1997). Saprobiological Water Quality Inestigations of the Sava River in Belgrade Region during 1996. 32. Arbeitstagung der IAD, SIL. Wissenschaftliche Kurzreferate, Wien: 397-402.

Karadžić, B. (2013). FLORA: a software package for statistical analysis of ecological data. Water Research and Management, 3, 45-54.

Karadzić, B., Sašo Jovanović, V., Jovanović, Z. and Popović, R. (1998). "Flora" A Database and Software for Floristic and Vegetation Analyzes. In Progress in Botanical Research (pp. 69-72). Springer Netherlands.

Kračun, M., Ilić, M., Tomović, J., Atanacković, A., Zorić, K., Vasiljević, B., Tubić, B., Marković, V. and Paunović M. (2013). Ocena stanja manjih vodotoka i kanala na teritoriji grada Beograda na osnovu zajednice makrobeskičmenjaka. „Zaštita voda 2013", Zbornik radova, Perućac, 53-58 pp.

Marković. V., Tomović, J., Kračun, M., Ilić, M., Zorić, K. , Vasiljević, B., Atanacković, A. , Tubić, B. and Paunović M. (2013). Freshwater Snails in the Belgrade Region. Ecological Truth XXI, Conference Proceedings, Bor, 296-301 pp.

Martinović-Vitanović, V., Kalafatić, V., Martinović, J. M., Paunović, M. and Jakovčev, D. (1999). Saprobiological analysis of benthic communities in the Danube in Belgrade region. Special issues of the Macedonian Ecological Society, 5: 504-516.

Official Gazette of the R. of Serbia 74/2011. The parameters of ecological and chemical status of surface waters and parameters of the chemical and quantitative status of groundwaters.

Official Gazette of the R. of Serbia 96/2010. Regulation on establishment of surface and groundwater bodies.

Paunović, M. M., Borković, S. S., Pavlović, S. Z., Saičić, Z. S. and Cakić, P. D. (2008). Results of the 2006 Sava survey: Aquatic macroinvertebrates. Archives of Biological Sciences, 60 (2): 265-271.

Paunović, M. M., Jakovcev-Todorović, D. G., Simić, V. M., Stojanović, B. D. and Cakić, P. D. (2007). Macroinvertebrates along the Serbian section of the Danube River (stream km 1429–925). Biologia, 62 (2): 214-221.

Pickett, S. T. A., Cadenasso, M. L., Grove, J. M., Boone, C. G., Groffman, P. M., Irwin, E., ... & Warren, P. (2011). Urban ecological systems: Scientific foundations and a decade of progress. Journal of Environmental Management, 92 (3): 331-362.

Popović, N., Jovanović, V., Raković, M., Kalafatić, V. and Martinović-Vitanović, V. (2013). Bottom Fauna Qualitative Study of the Danube River in Belgrade Region. Acta Zoologica Bulgarica, 65 (4): 505-516.

Smith, R.F. andLamp,W.O. (2008). Comparison of insect communities between adjacent headwater and main-stem streams in urban and rural watersheds.Journal of North American Benthological Society 27 (1): 161-175.

Vermonden, K. (2010). Key factors for biodiversity of urban water systems.. PhD-thesis, Radboud University, Nijmegen,147pp

Walsh, C. J., Roy, A. H., Feminella, J. W., Cottingham, P. D., Groffman, P. M. and Morgan, R. P. (2005). The urban stream syndrome: current knowledge and the search for a cure. Journal of the North American Benthological Society, 24 (3): 706-723.

 

 

The Distribution of Astacidae (Decapoda) Fauna in Kosovo and Metohija, Serbia

Nebojša V. Živić1, Ana Atanacković2, Slaviša Milošević1, Maja Milosavljević1

 

 

1 University of Priština/ Kosovska Mitrovica, Faculty of Science and Mathematics, Kosovska Mitrovica, Serbia; E-mail: This e-mail address is being protected from spambots. You need JavaScript enabled to view it

2 Institute for Biological Research “Siniša Stanković”, University of Belgrade

 

 

Abstract

Extensive regulatory monitoring programs have been conducted in Europe aiming at the protection and risk assessment of crayfish. In this sense, the paper presents data on the distribution of crayfish in Kosovo and Metohija. We analyzed data from the period between 1992-2012. Two of the four species reported for the Republic of Serbia were found in Kosovo and Metohija: Austropotamobius astacus and Astacus torrentium. A total of 33 sites were investigated that have swept all three sea watersheds: the Adriatic, Aegean and Black Sea. The species A. astacus prefers lowland water. The species A. torrentium inhabits mountainous ecosystems with water of oligo- to beta- mezosaprobic level of water quality. The site altitude of distribution of the A. torrentium species was at over 700m. Lowland parts of the investigated rivers were populated with individuals of A.astacus.

Keywords: Crayfish, distribution, Kosovo and Metohija, Serbia.

 

Introduction

Many studies were carried out on the distribution of the species of the family Astacidae in Europe (Holdich, 2002a) and other parts of the world (Brodsky, 1977; Starobogatov, 1995). Studies of freshwater crayfish have given good results in various explored approaches: commercial (Skurdal & Taugbøl, 2002), economic (Laurent, 1988) and morphological (Holdich, 2002b), evolutionary and phylogenetic (Scholtz, 2002), anatomical (Vogt, 2002), growth and reproduction (Reynolds, 2002), ecological (Nyström, 2002), taxonomic and protection of native species (Taylor, 2002), behavior (Gherardi, 2002), genetic variability (Fetzner & Crandall, 2002), physiological (McMalon, 2002), agents of diseases, parasites and commensals (Evans & Edgerton, 2002), immunological (Söderhal & Söderhal, 2002), as well as on the analysis of DNA (Huber & Shubert, 2004). The European continent is inhabited by five native freshwater crayfish species from the Astacidae family: Astacus astacus (Linnaeus, 1758), river or noble crayfish, Astacus leptodactilus Eschscholtz 1823, Danube, Turkish or swamp crayfish, Astacus pachypus Rathke 1837, Austropotamobius pallipes (Lereboullet 1858) white-leg or coastal crab and Austropotamobius torrentium (Schrank, 1803) rocky crayfish or steam crayfish (Holdich et al., 1999).

All native species from the genus Astacus Pallas 1772 and Austropotamobius Skorikow 1908 are listed in the Red List (IUCN, 2010). In European countries, native species are differently distributed. Four native species from the family Astacidae: A. astacus, A. leptodactylus, A. torrentium and A. pallipes are characteristic for the area of ​​the Balkans (Obradović, Ј. 1984; Maguire et al., 2004 Karaman, 1976; Bedjanič, M. 2004; Simic et al., 2008; Rajkovic, 2007; Trožić-Borovac, 2011). In Serbia, the presence of three native species A. astacus, A. leptodactylus, A. torrentium is recorded, The status of the species A. astacus and A. torrentium according to IUCN criteria (IUCN, 2010), was designated as "Lower Risk / Near Threatened" ("LR / nt"), Which is lower than the international level of this species "Vulnerable" (Simić, et al., 2008). Studies of freshwater crayfish fauna in Kosovo and Metohija have so far been rare, and as a part of the environmental watercources studies (Živić, 1998). This work was undertaken with the aim to show the presence of the species from the family Astacide in Kosovo and Metohija and to contribute to the knowledge of their distribution in order to protect them in the Balkans and Europe.

Material and Methods

Data on the distribution of crayfish species in Kosovo and Metohija originated from a number of studies carried out in the period 1992-2012. These researches implicated sites in the aquatic ecosystems of three catchments: the Adriatic, Aegean and the Black Sea. Crayfish were collected using different nets, improvised traps, as well as by hand capture. The numerous sites have been investigated, but only the sites where representatives of freshwater crayfish were found are presented in this work. During the sampling, in order to register the presence of crayfish species, a maximum of six individuals were collected, preferably both sexes of the same species. Other captured individuals were returned into their habitat. Representatives of crayfish species were found at the 33 sites tested. Determination of caught animals was performed according to the keys for the family Astacidae (Bott, 1972; Karaman, 1961, 1963; Holdich et al., 2006; Parvulescu, 2008; Zaikov, 2010). The territory of Kosovo and Metohija covers an area of ​​10,887 km2 and lies between 41°50'58" and 43°15'42" north latitude and 20°01'30" and 21°43'10" east longitude. This area has a relatively poor surface hydrographic network and its lack of water, especially during the summer. The problem of dense human population is reflected in the lack of water for drinking and hygiene needs. Water flows are burdened with wastewater of industrial and urban origin.

Results and Discussion

Distribution of river crayfish fauna has its cause-effect relationships. From their centers of origin, the Ponto-Caspian Basin, the freshwater craayfish population from the family of Astacidae is expanding their distribution area throughout Europe (Karaman, 1962, 1963). In the long process of adjustment many species have become native to certain areas. In recent years, native crayfish species are in a constant struggle for their survival. Violation of the physical and chemical characteristics of the ecosystem, intentional introduction of alien species or the appearance of invasive potential species has impact on reducing the populations of native crayfish species (Gherardi and Holdich, 1999, Taylor, 2002). Knowing the distribution of crayfish species from the family Astacide contributes to their conservation and survival. Out of the five native crayfish species that live in fresh water in Europe, three are registered in the Republic of Serbia: A. astacus, A. leptodactilus and A. torrentium. Distribution of freshwater crayfish species is poorly researched on the territory of Kosovo and Metohija. A partial view of the river crayfish fauna of this area was encountered in the study of epibionts from the family Branchiobdelidae (Annelida, Oligochaeta) (Karaman, 1967), and Hydrobiological study of the Sitnica River basin (Živić, 1998). During long term research, two crayfish species were recorded: A. astacus (Linnaeus, 1758), river or noble crayfish and A. torrentium (Schrank, 1803), rocky cancer or steam cancer inhabited watercources in Kosovo and Metohija. Both species are listed in the Red List (IUCN, 2010) and their habitats in the Habitat Directive. Investigated sites mainly include the entire territory of Kosovo and Metohija. The presence of individuals of these species was recorded at 33 sites. A. torrentium was found in 14, while A. astacus was found at 19 sites (Tab. 1).

 

Tab01

The conducted study did not have a quantitative approach, so the number of individuals found was not relevant. The history of collection of crayfish species in this area is long-lasting. In the period between 1982-1996, the author and his associates caught dozens of individuals A.astacus for lab exercises with students, in the river Nerodimka at the site of Donja Grlica village. Some specimens have weighed up to 300 g, and have been up to 30 cm in length. By inflow of waste water from Uroševac Town in the Nerodimka River, the A.astacus population withdrew to the upstream parts of the river. The expansion of A. torrentium in hydrogeographic networks of Europe was suppressed by the phylogenetically younger population of species A.astacus, inhabiting waters of Eastern Europe and the Balkan Peninsula. Suppressed from the lowlands the A. torrentium species retreated into mountain streams and rivers. The adaptive plasticity of A. torrentium enabled it to adapt to high mountain river ecosystems. This species was found in the source branches of all three hydrographic basins: the Adriatic, Aegean and the Black Sea. This zonation in the species distribution of the populations of A.astacus and A. torrentium registered in streams in the region of Kosovo and Metohija which was in accordance with similar habitats in the neighboring countries (Rajkovic, 2007; Parvulescu, 2008; Zaikov, 2010; Trožić-Borovac, 2011). It was found that species A. torrentium inhabits mountain streams and small rivers, the source branches of lowland rivers. Water rich in oxygen, moderate to low temperatures and rocky bottom are the characteristics of these habitats. It was found that these watercourses were characterized by an oligo saprobic level of water quality. The site altitude of the distribution of this species was to over 700m. Lowland parts of the investigated rivers were populated with individuals of A.astacus.

The most important habitat of this species was a hydrographic basin of the Sitnica River, the major river of the Kosovo Valley (Fig. 1).

Based on previus research, the water quality of finding sites of A. Astacus was assessed as beta-mezosaprobic (Urošević, 1989; Živić, 1997, 1998). Large expansion of the human population is evident at the investigated area. It is followed by a large urban, industrial and agro-technical change. Therefore, crayfish have reduced and fragmented areals. A.astacus native species and A. torrentium are listed in the Red List (IUCN, 2010) in Europe while their habitats in the Habitat Directive. Therefore, ecosystems that they inhabit in Kosovo require a special conservation approach and research to ensure their protection and biodiversity conservation, similar to other regions in Europe (Fyreder at al., 2004).

 

Conclusion

The undertaken research has shown that the water bodies in the territory of Kosovo and Metohija are populated by native populations of the two species of freshwater crayfish – Astacus astacus and Austropotamobius torrentium. Their presence was confirmed at 33 sites. Individuals of A. torrentium inhabit mountain streams and rivers of oligosaprobic water quality. Populations of A. astacus occupy lowland parts of the rivers with a beta meszosaprobic level of water quality. Site altitude of the distribution of this A. torrentium species was to over 700m. Lowland parts of the investigated rivers were populated with individuals of A. astacus.

 

Fig01
Figure  1: Hydrographic network of crayfish distribution.

References

Bedjanič, M. 2004: Novi podatki o razširjenosti raka navadnega koščaka Austropotamobius torrentium (Schrank, 1803) v Sloveniji (Crustacea: Decapoda) Natura Sloveniae 6(1): 25-33.

Bott, R. 1972: Besiedlungsgeschichte und Systematik der Astaciden West-Europas unter besonderer Berücksichtigung der Schweiz. Revue Suisse de zoologie 79 (13): 387-408.

Brodsky, S.Y. 1977: River crayfishes (Crustacea, Astacidae) of the Soviet Union. Vyestnik zoologii 3:48-53.

Evans, LH., Edgerton, BF. 2002: Pathogens, parasites and commensals. U: Holdich DM (ed.) Biology of freshwater crayfish. Bleckwell science, Oxford, 377-438.

Fetzner, JW., & Crandall, K.A. 2002: Genetic variation. U: Holdich DM (ed). Biology of freshwater crayfish. Bleckwell science, Oxford, 291-327.

Fyreder, L., Sint, D., Leiter, J., Declara, A. 2004: Species protection programs on autochthonous crayfish in Tyrol, Austria and Italy. Institute of Zoology and Limnology, University of Innsbruck, Innsbruc.

Gherardi, F., Holdich, DM. 1999: Crayfish in Europe as alien species. How to make the best of a bad situation? AA Balkema, Roterdam, Brookfield.

Gherardi, F. 2002: Behaviour. U: Holdich DM (ed.) Biology of freshwater crayfish. Bleckwell science, Oxford, 258-291.

Grupa autora: 1996: Elaborat - Ribarsko područje "Kosovo i Metohija I - II",.Republičko Ministarstvo za prostorno planiranje Republike Srbije (ne publikovani materijal), 1-41.

Holdich, DM., Ackefors, H., Gherardi, F., Rogers, WD., Skurdal, J. 1999: Native and alien crayfish in Europe: Some conclusions. U: Gherardi, F. I Holdich, DM. (eds.) Crayfish in Europe as alien species. How to make the best ofa bad situation? AA Balkema, Roterdam, Brookfield, 281-292.

Holdich, DM. 2002a: Present distribution of crayfish in Europe and some adjoining countries. Bull.Francais de la Peche et de la Pisciculture367, 4,611-650.

Holdich, DM. 2002b: Beckground and functional morphology. U: Holdich (ed.) Biology of freswater crayfish. Blackwell Science, Oxford, 1-27.

Holdich, D. & Vigneux, E. 2006: Key to crayfish in Europe. In: Soute-Grosset, C., Holdich D., Noel P., Reynolds J. & Haffner P., (eds): Atlas of crayfish in Europe. Museum National d'Histoire naturelle, Paris, 53-55.

Huber, M. G. J., Schubart, C. D. 2004: Genetic analysis of the stone Crayfish (Austropotamobius torrentium) in Germany on the local distribution around Regensburg and the impact of alien crayfish species. Universität Regensburg Leehrstuhl Zoologie, Regensburg. IUCN, 2010: IUCNRed List of Threatened Species. Version 2010.1. . Кaraman, М.S. 1976: Fauna na Makedonija. Desetonogi rakovi-Decapoda (Malacostraca –Crustacea).Prirodonaucen muzej –Skopje.

Кaraman, М S. 1961: Slatkovodni rakovi Jugoslavije. Ribarstvo Jugoslavije 3 (61):1-33. Кaraman, М. S. 1962: Ein Beitrag zur Systematik Astacidae (Decapoda). Crustaceana 3: 173-191.

Кaraman, М.S. 1963: Studie der Astacidae (Crustacea, Decapoda) II. Teil. Hydrobiol. 22: 111-132.

Karaman, M. S. 1967: Branchiobdellidae Jugoslavije. – Zbornik filozofskog fakulteta u Prištini, IV: 39-64.

Laurent, P.J. 1988: Austropotamobius pallipes and A. torrentium, with observations on their interactions with other species in Europe. U Holdich DM i Lowery RS (eds.) Freshwater crayfish. University Press, Cambridge 341-364.

Maguire, I. and Gottstein-Matocec, S. 2004: The distribution pattern freshwater crayfish Croatia by. Crustaceana 77(1) : 25-49.

McMalon, B.R. 2002: Physiological adaptation to environment: U: Holdich DM (ed) Biology of freshwater crayfish. Bleckwell science, Oxford,327- 377.

Nystrom, P. 2002: Ecology.U: HoldichDM(ed) Biology freshwater crayfish. Bleckwell science, Oxford, 192-236.

Obradović, Ј. 1984: Problem terminologije slatkovodnih rakova. Bilten Društva Ekologa Bosne i Hercegovine. Serija B-Naučni skupovi I savjetovanja. Broj 3- III Kongres ekologa Jugoslavije, knjiga II: 165-169.

Pârvulescu, L., 2010: Crayfish field guide of Romania. Editura Bioflux Ceahlău St. No. 54 RO 400488, Cluj-Napoca. www.crayfish.ro

Rajkovic, М. 2007: Održivo korišćenje populacije riječnog raka Astacus astacus (Linnaeus 1758) u vodenim ekosistemima gornjeg toka rijeke Zete. Magistarski rad. Univerzitet u Kragujevcu.

Reynolds, J.D. 2002: Growth and reproduction. U: Holdich DM (ed.) Biology of freshwater crayfish. Bleckwell science, Oxford, 152-192.

Scholtz, G. 2002: Phylogeny and evolution. U: Holdich DM (ed) Biology of freshwater crayfish.Bleckwell science, Oxford, 30-52.

Simić, V.,Petrović, А., Rajković, М., Paunović, М. 2008: Crayfish of Serbia and Montenegro – the population status and the level of endangerment. Crustaceana 81 (10): 1153-1176.

Skurdal, J. & Taugbol, T. 2002: Astacus. In Holdich, D. M. (ed.), Biology of Freshwater Crayfish. Blackwell Science, 467-510, Oxford.

Söderhäll, I. & Söderhäll, K., 2002: Immune reactions. U: Holdich DM (ed.) Biology of freshwater crayfish. Bleckwell science, Oxford, 439-464.

Starobogatov, Yal. 1995: Taxonomy and geographical distribution of crayfish of Asia and East Europe ( Crustacea, Decapoda, Astacoidei). Arthropoda Selecta 4(3): 3-25.

Taylor, C.A., 2002: Taxonomy and conservation of native crayfish stocks. U: Holdich, DM (ed) Biology of freswater crayfish. Bleckwell science, Oxford, 236-257.

Trožic-Borovac, С. 2011: Freshwater crayfish in Bosnia and Herzegovina: the first report on their distribution Knowledge and Management of Aquatic Ecosystems (2011) 401, 26.

Urosevic, V. 1989: Supplement to the study of the algae of the rivers Sitnica and Ibar, Acta Biology and Medicus Experimental, University of Pristina 14, 117-127.

Vogt, G. 2002: Functional anatomy. U: Holdich DM (ed.) Biology of freshwater crayfish. Bleckwell science, Oxford, 53-151.

Zaikov, A. 2010: An illustrated key to the Bulgarian freshwater crayfish species of family Astacidae (Crustacea: Decapoda) ZooNotes 13: 1-4

Živić, V.N., Šapkarev, J., LAbus, N. 1997: Composition and Distruution of Leeches (Annelida: Hirudinea) in River Sitnica Basin and in River Ibar. University Thought. Publication in Nature Science, vol.4, No.2, 97-103. Pristina.

Živić, V. N., 1998: Macrozoobenthos community as biological indicator of ecological condition of Sitnica river basin. Doctoral thesis University of Novi Sad, Faculty of Natural Scineces and Mathematics, 1-143.