Thermal and Dissolved Oxygen Properties and Fish Assemblages of the Zlatar Reservoir


Branislav Mićković1, Miroslav Nikčević1, Vesna Djikanović2, Marija Smederevac-Lalić1, Zoran Gačić1 and Aleksandar Hegediš1




1 Institute for Multidisciplinary Research, University of Belgrade, Kneza Višeslava 1, Belgrade, 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, Bulevar Despota Stefana 142, Belgrade, Serbia




The aim of the study was to describe thermal and oxygen characteristics of the reservoir and to determine long-term changes in the fish community. Zlatar reservoir represents a typical dimictic type of lake with clearly developed summer stratification and differentiated layers of epilimnion, metalimnion and hypolimnion, whereas isothermal conditions occur in spring and autumn. Throughout the annual cycle reservoir water is was well-oxygenated, whereas in summer, hypersaturated oxygen conditions in the epilimnion and metalimnion were observed. The species community currently present contains 10 species but only 4 are native. Results showed that among typically rheophilic species only a few are generalist enough to remain competitive in a lacustrine environment. The bleak and Danube roach were dominant species in terms of numbers with the latter being the dominant for biomass, whereas chub represented the subdominant species in terms of abundance and biomass.

Keywords: temperature, oxygen, stratification, Zlatar Reservoir




River impoundments (dam reservoirs) belong to the group of the most frequent types of heavily modified waters, with substantial ecological, economic and recreational importance. Although apparently similar, compared to natural lake reservoirs, they exhibit numerous functional differences that hinder their comparison (Wetzel, 1990). Reservoirs can be regarded as particularly interesting environments due to their youth, namely, their rapid evolution in comparison to natural hydrosystems providing valuable information on the ecological successions following a major environmental disruption (Irz et al. 2002).

In Serbia there are numerous reservoirs in use and more are being planned or under construction. With a volume of approximately 250 million m3, Zlatar Reservoir is among the biggest reservoirs in Serbia and is considered to be the second largest storage for drinking water production in Serbia. Moreover, it is in a protected area (Special Natural Reserve Uvac) which is recognized as a territory with high significance for conservation of biodiversity in Serbia. Despite these facts its environmental conditions and ichthyofaunistic characteristics are not properly investigated and documented. In this study our objectives are: 1) to describe seasonal thermal and dissolved oxygen properties of the reservoir as the environmental parameters of the highest biological and fisheries significance; 2) to determine long-term changes in the fish assemblages; 3) to provide necessary background for the assessment of reservoir fish assemblages which can be useful for proper management of fish communities in the protected water body.

Material and Methods

The Zlatar reservoir occupies an intermediate position in the chain of three river impoundments built on the middle course of the Uvac River (Fig. 1). The earthen dam (1,264 m in length, 83 m in height) was built in 1962 at an altitude of 880 m.a.s.l., creating an artificial lake of 7.25 km2 surface area and a maximum depth of approximately 75 meters. The reservoir receives water from the above-located Uvac Reservoir and has a pronounced canyon-shaped morphometry with three small tributaries.

Surveys of thermal and dissolved oxygen properties of the Zlatar Reservoir were conducted in August 2011, October 2011, and April 2012. In August and October measurements were performed on six locations along the longitudinal transects (Fig. 1), while in April measurements were performed only on location in the middle part of the reservoir. At each location, the vertical profiles of water temperature and dissolved oxygen concentrations were measured from 0 m to 22 m depth, with 2 m intervals. Water sampling was performed using the Ruttner water sampler of 5 L volume and measurements of temperature and dissolved oxygen concentration were done with a Windaus field laboratory. Fish sampling was performed in July and October 2011 using benthic and pelagic gillnets and benthic trammel nets (mesh size 12-70 mm). Nets were set overnight and a total of 389 fishes from 10 different species caught measured (total length, ± 1 mm) and weighed (± 0.5 g). The results were presented as the abundance and biomass of all fishes and a four grade dominancy scale was applied: > 20% dominant (D), 10.1-20% subdominant (Sd), 1.1 -10% common (C) and < 1% rare (R). Non – native species to Uvac River catchment area were considered as introduced either being translocated or alien.


Figure 1: Study area; solid squares mark sampling sites


Results and Discussion

During both seasons, each depth strata was characterized by homogeneous values of temperature and concentration of dissolved oxygen along the longitudinal profile (Table 1). Recorded ranges for both measured indicators were narrow and respective values of coefficient of variability varied in ranges of 1.91-8.62% and 1.79-14.78% for temperature and oxygen concentrations recorded in August, and 0.41-10.43% and 3.01-12.80% for temperature and oxygen concentrations recorded in October.

Zlatar reservoir is a large, deep, canyon-shaped water body. In physically complex reservoirs, such as the one studied, the degree of dendricity and sinuosity arising from river impoundment can result in distinct differences in water physical and chemical characteristics over the length of an impounded river course. In these instances, longitudinal zonation of water quality may occur and a single sampling site is not sufficient to provide reliable data for adequate assessment of environmental conditions for the entire water body. The precise knowledge of the temperature and dissolved oxygen horizontal and vertical distribution patterns under different meteorological, physical and biological conditions represent a valuable tool in the hands of reservoir managers (Moreno-Ostos et al., 2006). It is essential for determination of the minimum number of sampling stations and depths required for environmental conditions and biota monitoring surveys. According to that, based on the selected sites, reservoir managers can design monitoring procedures in order to minimize time and cost.

Additionally, the integration of this type of data in dynamic reservoir modeling would improve the quality and precision of simulations and predictions, thus advancing the application of these kinds of tools for scientifically based reservoir management (Moreno-Ostos et al., 2006). Our results showed, despite the complex morphology of the Zlatar reservoir, the existence of a relatively homogeneous pattern of longitudinal distribution of the measured indicators for certain depths along the entire reservoir. Therefore, it seems that a single sampling site with a reduced number of vertical profiles would be sufficient to explain the seasonal aspect of thermal and dissolved oxygen characteristics of the reservoir. However, the single sampling site approach should be considered only as an operational approach and it should not be used in situations where a scientific and statistically valid assessment of water quality is required.

Vertical profiles of temperature and dissolved oxygen concentrations recorded along the investigated water column are presented in Figure 2. Thermal and dissolved oxygen profiles recorded in August and October refer on average values (Table 2), while the profiles recorded in April were obtained from one sampling site procedure. Absolute thermal amplitudes were 14.84 °C, 2.6 °C and 1.4 °C for profiles recorded in August, October and April, respectively. As shown in Figure 2, in August the water column exhibited clear stratification with a thermocline between depths of 6 m (20.92 °C) and 12 m (11.63 °C). In October and April, the thermal gradients were essentially uniform along the water column, exhibiting an orthograde profile.


Table 1: Seasonal ranges and average values of water temperature recorded along the longitudinal transects (six sampling locations) in respect to the investigated depths.