Acta Botanica 2-2016 - za web.indd ACTA BOT. CROAT. 75 (2), 2016 199 Acta Bot. Croat. 75 (2), 199–205, 2016 CODEN: ABCRA 25 DOI: 10.1515/botcro-2016-0032 ISSN 0365-0588 eISSN 1847-8476 The application of benthic diatoms in water quality assessment (Mlava River, Serbia) Olga S. Jakovljević1*, Slađana S. Popović1, Danijela P. Vidaković1, Katarina Z. Stojanović2, Jelena Ž. Krizmanić1 1 University of Belgrade, Faculty of Biology, Institute of Botany and Botanical Garden »Jevremovac«, Takovska 43, 11000 Belgrade, Serbia 2 University of Belgrade, Faculty of Biology, Institute of Zoology, Student Square 16, 11 000 Belgrade, Serbia Abstract–The main objective of this study was to assess the ecological status of the Mlava River based on epilithic diatoms and to test the use of diatom indices as a tool for estimating the quality of fl owing waters in Serbia. Quantitative analysis showed that in April Achnanthidium minutissimum was dominant at each site, except at the fi fth site, where Amphora pediculus was dominant. In July and September, Achnanthidium minutissimum, Achnanthidium pyrenaicum, Amphora pediculus, Denticula tenuis, Diatoma vulgaris, Gom- phonema elegantissimum, Cocconeis pseudolineata and Cocconeis placentula var. lineata dominated. De- trended correspondence analysis (DCA) was used to detect the major patterns of variation in species compo- sition. The fi rst DCA axis summarizes the distribution of the diatom community, mainly through temperature, conductivity, oxygen and water hardness gradient. The second DCA axis was weakly correlated with few variables. Based on the average values of the pollution sensitivity index (IPS), commission for economical community metric (CEE) and biological diatom index (IBD), the water of the Mlava River belonged to water class I during all three seasons. Values of the diatom-based eutrophication/pollution index (EPI-D) indicated class II water quality. According to calculated trophic diatom index (TDI) values, water of the Mlava River was characterized by intermediate nutrient concentrations during three seasons. Principal components analy- sis was used to represent the correlation between diatom indices, and the highest correlation among the se- lected diatom indices is seen between EPI-D, IPS and IBD. Key words: biomonitoring, diatoms, indices, Mlava River * Corresponding author, e-mail: olga.jakovljevic@bio.bg.ac.rs Introduction The best way to evaluate the ecological status of waters is the simultaneous use of physico-chemical and biological analysis. Use of only physico-chemical methods gives un- reliable results, because water quality may change over a short period of time. The really current state of the ecosys- tem can only be obtained by using biological methods. Ac- cording to the Water Framework Directive (WFD 2000/60/ EC), biological indicators play a key role in the assessment of ecological status. Biological assessment is expressed ac- cording to a numerical scale between zero and one – the ‘ecological quality ratio’ (EQR). Values of this scale ranged from zero to one, with high status represented by values close to one and bad status by values close to zero. ‘Macro- phytes and phytobenthos‘ are together one of the biological quality elements (BQEs) for assessment of the ecological status of European rivers and lakes. In total, 66 macrophyte and phytobenthos assessment systems (32 for macrophytes, 30 for phytobenthos and 4 combined) have been developed and intercalibrated (Poikane et al. 2016). When it comes to phytobenthos, benthic diatoms are a valuable tool in water quality assessment and monitoring because they occur in most aquatic ecosystems, at any time, are easy to collect and sensitive to physical, chemical and biological changes in the water (Vasiljević et al. 2014). However, there are some diffi culties in the use of diatoms in water quality as- sessment. One disadvantage is the large number of taxa in- volved, which is partly resolved by using an index based on genera (Rumeau and Coste 1988). Also, considerable taxo- nomic expertise is required (Ács et al. 2004). In addition, pH, conductivity, total phosphorus, temperature, alkalinity, altitude, nitrates, calcium, biological oxygen demand (BOD), chlorophyll a and substrate type are major environ- mental factors that affect diatom distribution in streams, in addition to light availability, total phosphorus and grazing pressure (Toman et al. 2014). JAKOVLJEVIĆ O. S., POPOVIĆ S. S., VIDAKOVIĆ D. P., STOJANOVIĆ K. Z., KRIZMANIĆ J. Ž. 200 ACTA BOT. CROAT. 75 (2), 2016 Diatoms have been used for water evaluation purposes for more than a decade in many countries of Europe (Aus- tria, Switzerland, Germany, Belgium, France, Poland, Fin- land, Luxemburg, United Kingdom, Spain, Portugal, Italy). According with the requirements of the WFD, the epilithic diatoms were investigated for ecological status assessment of the Aksu and Isparta streams, as well as the Porsuk and Karasu rivers in Turkey (Solak 2011). Ecological status as- sessment, based on diatom indices, is still a new topic in Turkey and the number of studies on this topic is constantly on the increase (Solak and Àcs 2011). Diatom indices have been widely used for Polish rivers and streams. Algological investigation of 38 Polish rivers was conducted and it was found that generic diatom index and pollution sensitivity index (IPS) indices are the most appropriate for Polish con- ditions. Many rivers were characterized by high and good ecological status (Solak 2011). Diatom studies have been sporadic and focused on large rivers in Hungary. Ács et al. (2009) investigated 398 streams and found that more than half had good ecological status based on the values of IPS, Austrian saprobic index and Austrian trophic index. The implementation of EU standards has been started and these countries possess information on ecological status of their most important rivers (Solak 2011). The Offi cial Gazette of the Republic of Serbia (»Sl. glasnik RS«, No. 74/2011) prescribes the use of two diatom indices: IPS (Coste in Cemagref 1982) and commission for economical community metric (CEE) (Descy and Coste 1991). The WFD has not been suffi ciently integrated in Ser- bia. Examination of diatom indices applicability in the as- sessment of the ecological status of fl owing waters in Ser- bia started during 2012 (Andrejić 2012). In that study, the trophic diatom index (TDI) was used in assessment of the water quality of the Nišava River and its tributaries the Jer- ma and Temska. After that, the ecological status of the Djetinja River was determined using the diatom pollution index (DAIpo) (Krizmanić et al. 2013). In all, 17 diatom indices were calculated with the help of a software package OMNIDIA (Lecointe et al. 1993) based on which the quali- ty of the Raška River was estimated (Vidaković 2013). Wa- ter quality assessment of the study area of the Danube- Tisza-Danube (DTD) hydrosystem has been evaluated by four diatom indices (Watanabe’s index-DAIpo, biological diatom index-IBD, TDI and IPS) (Jakovljević et al. 2014). Serbia is in the early stage of preparation in the fi eld of environment and climate changes (Denić et al. 2015). It is necessary to audit the »Ordinance on the parameters of the ecological and chemical status of surface waters and the pa- rameters of the chemical and quantitative status of ground- water«, (»Sl. glasnik RS«, br. 74/2011) in the part of the parameters list and class boundaries for individual parame- ters of the biological quality elements. Based on the catego- ry of ecological status expressed through ecological quality ratios (EQR) values (WFD CIS Guidance No. 7. 2003), ex- cellent ecological status is achieved when the EQR value is greater than 0.80. In all, 6 samples over 3 years are needed for reliable ecological status classifi cation (WFD-UKTAG 2014). The aim of this study was to assess the ecological status of the Mlava River based on epilithic diatoms and to test the use of diatom indices as tool for estimating the quality of fl owing waters in Serbia. Materials and methods The Mlava River is located in the northwestern part of Eastern Serbia with a course of 120 km, covering a drain- age area of 1830 km2, and fl ows into the Danube. It is one of the longest rivers in Eastern Serbia. The Mlava River is particularly specifi c in physical-geographical characteris- tics (geological structure of the terrain and water level of the basin) (Manojlović et al. 2012). The upper part of the river fl ows through a gorge and the downstream part through a wide valley (Babić Mladenović 2009). A trout fi sh pond was created in the upper part of the investigated section of the Mlava River, between the fi rst and second sampling site. Some basic data about the sites are given in Tab. 1. The sampling was conducted during three seasons (April, July and September 2011) from 5 localities (ML1- ML5), along the Mlava River. Benthic diatoms were col- lected from the stones by scraping with a toothbrush. The material was preserved in 4% formalin. Water temperature, conductivity, pH, ammonium ions, nitrates, oxygen, bio- chemical oxygen demand, alkalinity, total phosphorus, or- thophosphates and water hardness were also determined in the water samples from each sampling site. Diatom frustules were cleaned using the standard meth- od with concentrated sulfuric acid (H2SO4), potassium per- manganate (KMnO4) and oxalic acid to remove the organic content (Krammer and Lange-Bertalot 1986). When pH was approximately 7, the material was mounted in Naphrax® mounting medium and permanent slides were made. Micro- scopic analysis of the permanent slides was performed using a Zeiss Axio-Imager M1 microscope with a digital camera AxioCam MRc5 and AxioVision 4.8 software. Tab. 1. Characteristics of the 5 investigated sites (ML1-ML5), along the Mlava River (Serbia). Site ML1 ML2 ML3 ML4 ML5 Altitude [m] 314 311 310 297 296 Water depth [m] 0.29 0.17 0.2 0.16 0.23 Flow width [m] 12 7.27 9.63 6.81 7.16 Flow velocity [m s–1] 0.1 0.44 0.39 0.65 0.37 Watercourse bottom gravel: 40% stones: 75% stones: 75% stones: 70% stones: 75% Coordinates 44°11.493N 021°47.012E 44°11.842N 021°46.516E 44°11.781N 021°46.158E 44°12.133N 021°45.131E 44°13.488N 021°44.635E WATER QUALITY MONITORING USING BENTHIC DIATOMS ACTA BOT. CROAT. 75 (2), 2016 201 Relative abundance of each diatom taxa was determined by counting 400 valves in each slide. Seventeen diatom in- dices are calculated based on the indicator values of identi- fi ed taxa using the OMNIDIA software (Lecointe et al. 1993), and fi ve were taken into consideration for the water quality assessment. These were: IPS (Coste in Cemagref 1982), eutrophication and/or pollution index – diatom based – EPI-D (Dell’Uomo 2004), CEE (Descy and Coste 1991), IBD (Coste et al. 2009) and TDI (Kelly and Whitton 1995). IPS, EPI-D, CEE and IBD are scaled from 1 to 20, while TDI is scaled from 1 to 100. Higher values of the fi rst four indices indicate water quality improvement while higher values of TDI indicate bigger trophy of water (Tab. 2) (Le- cointe et al. 1993). Statistical analyses, detrended correspondence analysis and principal component analysis, were done using CANO- CO for Windows Version 5 (Ter Braak and Šmilauer 2012). Results Qualitative and quantitative analysis of diatoms An investigation of the benthic diatoms from the Mlava River during three seasons resulted in the description of 86 diatom taxa, belonging to 27 genera. The greatest taxa rich- ness was recorded within the following genera: Navicula Bory sp. (12), Gomphonema Ehrenberg sp. (11), and Nitz- schia Hassall sp. (8). Samples from the third sampling site (ML3) had the highest diversity of benthic diatoms in all seasons. Forty-fi ve taxa were recorded in April, 39 taxa in July and 30 in September. The lowest number of diatom taxa was recorded at the site ML1. In April, 16 taxa were recorded, 25 taxa in July and 17 in September. According to the frequency of the taxa, Amphora pediculus (Kützing) Grunow, Denticula tenuis Kützing, Cocconeis placentula var. lineata (Ehrenberg) van Heurck, Planothidium frequen- tissimum Lange-Bertalot, Reimeria sinuata (Gregory) Ko- ciolek & Stoermer, Rhoicosphenia abbreviata (C.Agardh) Lange-Bertalot and Encyonema minutum (Hilse) D.G.Mann were the taxa that occurred at the most investigated sites. Cymbella excisa Kützing, Surirella brebissonii var. kuetzin- gii Krammer and Lange-Bertalot, Nitzschia pusilla Grun- ow, Cymbella excisiformis Krammer and Fragilaria para- sitica var. parasitica (W. Smith) Grunow were documented at only one site (Fig. 1). Quantitative analysis showed that in April Achnanthid- ium minutissimum (Kützing) Czarnecki was dominant at each locality, except at the fi fth site, where Amphora pedic- ulus was dominant. Also, Planothidium frequentissimum, Gomphonema micropus Kützing, Achnanthidium pyrenai- cum (Hustedt) H.Kobayasi, Amphora pediculus and Reime- ria sinuata (Gregory) Kociolek & Stoermer were dominant taxa during April. A. minutissimum, A. pyrenaicum (Hustedt) H.Kobayasi, A. pediculus, Denticula tenuis, Diatoma vul- garis Bory, Gomphonema elegantissimum E.Reichardt & H.Lange-Bertalot, Cocconeis pseudolineata (Geitler) Lange - Bertalot and Cocconeis placentula var. lineata dominated in July and September. The percentage participation of dominant taxa was 10% or more at least in one site. Tab. 2. Class boundary limits for diatom indices. PS – pollution sensitivity index, EPI-D –diatom-based eutrophication/pollution index, CEE – commission for economical community metric, IBD – biological diatom index, TDI – trophic diatom index. Water quality class Ecological status IPS, EPI-D, CEE, IBD TDI Trophic status average % of taxa used for the index calculation I high 17–20 <20 oligotrophic 100 (IPS) II good 13–16 20–40 oligo-mesotrophic 73.91 (EPI-D) III moderate 9–12 40–60 mesotrophic 74.23 (CEE) IV poor 5–8 60–80 eutrophic 93.61 (IBD) V bad 1–4 80–100 hypertrophic 82.49 (TDI) Fig. 1. Light micrographs of diatom species that occurred at most investigated sites (1–7); species that occurred at only one site (8– 13). 1 – Cocconeis placentula var. lineata; 2 – Rhoicosphenia ab- breviata; 3 – Encyonema minutum; 4 – Planothidium frequentissi- mum; 5 – Reimeria sinuata; 6 – Denticula tenuis; 7 – Amphora pediculus; 8 – Cymbella excisa; 9 – Surirella brebissonii var. kue- tzingii; 10 – Cymbella excisiformis; 11 – Fragilaria parasitica var. parasitica; 12 – Nitzschia pusilla. Scale bar = 10 μm. JAKOVLJEVIĆ O. S., POPOVIĆ S. S., VIDAKOVIĆ D. P., STOJANOVIĆ K. Z., KRIZMANIĆ J. Ž. 202 ACTA BOT. CROAT. 75 (2), 2016 Diatom-environmental relationships The values of physico-chemical parameters of the Mla- va River are presented in Tab. 3. Detrended correspondence analysis (DCA) was used to detect the major patterns of variation in species composi- tion (Fig. 2). The eigenvalues of the fi rst and second axis were 0.7398 and 0.3210 respectively. The high eigenvalue of the fi rst axis indicates its good explanatory power. The fi rst DCA axis summarizes the distribution of the diatom community, mainly through temperature, conductivity, oxy- gen and water hardness gradient. Sites, sampling time and taxa from the left side of the ordination diagram have a ten- dency toward higher oxygen and water hardness, but lower temperature and conductivity level, while sites, sampling time and taxa from the right side of the diagram show the opposite. As for the second axis, it is more diffi cult to inter- pret, since there are several variables weakly correlated with it, for example nitrates and orthophosphates. Higher levels of orthophosphates correspond to the lower part of the ordination diagram, and higher levels of nitrates to the upper part of the diagram. Correlation coeffi cients of the selected variables and the two DCA axes are given in On- line Suppl. Tab. 1. It is also worth saying that the sampling site ML1 had the lowest values of ammonia and total phos- phorous and highest values of oxygen and water hardness. The Species Response curves that explain difference for the selected species distribution along the fi rst DCA axis using generalized linear model are shown in On-line Suppl. Fig. 1. Since the fi rst DCA axis shows the highest negative cor- relation with oxygen and highest positive correlation with temperature, species response curves of selected taxa show that Achnanthidium minutissimum, Planothidium frequen- tissimum and Gomphonema micropus are more abundant at higher oxygen levels, while Cocconeis placentula var. lin- eata and Cocconeis pseudolineata are more to be found at higher temperatures. Diatom indices and water quality The values of the fi ve diatom indices used indicated a similar ecological status of the water at the studied sites. PCA was used to represent the correlation between diatom indices, where the highest correlation is seen between EPI, IPS and IBD. Also, CEE is correlated with them. These three indices had the highest values at sampling site ML2 during September. CEE had the highest value at sampling site ML1 during September and TDI at sampling site ML5 during April (Fig. 3). Based on the average values of the IPS, CEE and IBD diatom index, water of the Mlava River belong to water class I during all three seasons, which corresponded to high ecological status. Values of the EPI-D index indicated class II water quality (good ecological status), although these values were on the border with values indicating class I wa- ter quality. According to calculated TDI values, water of the Mlava River is mesotrophic with moderate nutrient concen- trations and corresponds to water class III during all sea- Fig. 2. Detrended correspondence analysis (DCA) of dominant diatom taxa (ADPY – Achnanthidium pyrenaicum, GMIC – Gom- phonema micropus, PLFR – Planothidium frequentissimum, AMIN – Achnanthidium minutissimum, RSIN – Reimeria sinuata, APEN – Amphora pediculus, DTEN – Denticula tenuis, GELG – Gomphonema elegantissimum, DVUL – Diatoma vulgaris, CPLI – Cocconeis placentula var. lineata, COPL – Cocconeis pseu- dolineata) in the Mlava River (Serbia) in the ordination space of fi rst and second axis with locality (squares; ML1–ML5), sampling season (circles; Apr, Jul, Sep) and physico-chemical characteris- tics (oxygen level-O2, water hardness-WH, biochemical oxygen demand-BOD, orthophosphates-OP, ammonia-Amm, total phos- phorous-TP, pH, temperature-T, nitrates-NO3, conductivity-Cond and alkalinity-Alc) as a supplementary variable. Tab. 3. Chemical water parameters of the Mlava River at the studied sites (ML1–ML5). BOD – biological oxygen demand. Parameter ML1 ML2 ML3 ML4 ML5 Temperature [°C] 10.4–11.4 10.5–15.2 12.8–16.8 10.6–14.6 10.8–15 Conductivity [μS cm–1] 378–435 374–490 355–506 340–490 351–475 pH 7.14–7.38 7.53–7.8 7.62–7.95 7.94–8.12 7.97–8.04 Ammonia [mg L–1] <0.0189–0.0471 0.143–0.4556 0.246–0.379 0.1648–0.3396 0.0505–0.176 Nitrates [mg L–1] 6.85–7.5 6.1–8 6.8–8.2 6.8–8.8 3.5–8.8 Oxygen [mg L–1] 11.1–12.2 7.7–10.8 9.4–11.2 10.6–10.9 9.1–10.8 BOD [mg L–1] 1.65–5.1 1.4–5.7 1.55–5.25 2–5.4 2.25–5.15 Alkalinity [mmol L–1] 3.6–4 3.6–4.1 1.29–4 3.4–4.2 3.5–4.2 Total phosphates [mg P L–1] 0.0335–0.052 0.0544–0.0817 0.0389–0.0942 0.0252–0.1128 0.0282–0.1151 Orthophosphates [mg P L–1] 0.0192–0.036 0.033–0.0344 0.0275–0.039 0.0134–0.0468 0.0171–0.0392 Water hardness [ºdH] 11.4–12.06 1.27–12.16 1.29–12.06 1.3–12.27 1.3–12.15 WATER QUALITY MONITORING USING BENTHIC DIATOMS ACTA BOT. CROAT. 75 (2), 2016 203 sons. TDI values during the fi rst (April) and second season (July) were on the border with values indicating the high nutrient concentrations. Values of all fi ve indices increase in the second and third season (September) (On-line Suppl. Tab. 2). Discussion Some of the most signifi cant factors affecting the distri- bution of benthic diatoms in the rivers are the physico- chemical characteristics of water, type of substrate, water velocity, amount of light and presence of predators (Steven- son et al. 1996). The infl uence of the physico-chemical characteristics of water has been widely studied by many authors. Patrick (1973) recorded that acidic waters do not support an abundance of Bacillariophyceae, but in alkaline waters with pH above 8.0, their density is much higher. Some studies indicate that pH in the neutral range (7.0–8.0), like the values recorded in our study (7.14–8.12), supports a good population of diatoms (Kumar and Oommen 2011). Optimal temperatures for benthic diatoms are in the range from 25 to 30 °C with diversity declining at the tempera- tures above 30 °C. According to Rusanov et al. (2009), con- ductivity related to geology, as well as total phosphorus, were the most important variables related to changes in the diatom assemblage structure. Phosphates, nitrates and silica are generally considered the most important nutrients in primary production, and it is documented that diatoms reach their maximum growth rate at a concentration of total phosphorus by 0.5 mg L–1 (Chételator et al. 1999). The DCA performed showed that the highest correla- tions with the fi rst DCA axis were with temperature and oxygen. Even though the temperature range in our study was quite low (10.4–16.8), DCA showed that some species of the genus Cocconeis (Cocconeis placentula var. lineata and Cocconeis pseudolineata) have a tendency toward higher temperatures (they are placed at the right side of the ordination diagram). Andrejić (2012) reported that optimal temperatures for Cocconeis genus are > 25 °C. It is known that Cocconeis placentula var. lineata has a wide ecological range, especially with regard to the electrolyte content and the trophic situation, even found in poorer electrolyte, sili- cate-dominated streams (Hofmann et al. 2013). One of the indicators of oxygen-rich waters, A. minutissimum, (Noga et al. 2014) is one of the most frequent taxa in epilithic dia- tom communities in streams and rivers in general (Virtanen and Soininen 2011). It is a cosmopolitan species that has a wide ecological spectrum recorded in water bodies with the conditions from oligo- to eutrophic with a wide range of pH (4.3–9.2). There are records of numerous populations that develop in high mountain streams (Van Dam et al. 1994, Kawecka 2012). It is also one of the most common and most frequently noted diatoms in the Podkarpacie region, Poland, especially in the upper courses of rivers and streams (Noga 2012, Pajączek et al. 2012). This diatom requires a continuously high concentration of dissolved oxygen. Oth- er taxa requiring a fairly high oxygen level (above 75% saturation) (Van Dam et al. 1994) are Planothidium fre- quentissimum and Gomphonema micropus. DCA showed good correlation between all these three taxa and higher oxygen level, but also with higher water hardness. It is worth mentioning that Planothidium frequentissimum was an indicator of high levels of organic pollution and electro- lyte contents in rivers in the Northeast of Spain and in France (Tornés et al. 2007, Rimet 2009) and appeared in samples with a phosphates concentration up to 5 mg L–1 and nitrates concentration up to 35 mg L–1. In our samples, or- thophosphates concentration was much lower and nitrates concentration was similar and varied 3.5–8.8. Amphora pe- diculus is an alkaliphilous species, often found in waters with moderate conductivity and tolerant to increased con- centrations of organic nitrogen (Van Dam et al. 1994). It colonized oligosaprobic and mesosaprobic habitats (Hof- mann et al. 2013) corresponding to those of the studied sites of the Mlava River. Based on PCA analysis conducted by studying diatoms of the Nišava River, it was observed that A. pediculus is positively correlated with nutrients (Andrejić 2012). It can be seen from our study that this spe- cies is correlated with nitrates but also with conductivity. Achnanthidium pyrenaicum and Reimeria sinuata are placed on the opposite sides of the DCA second axis, but it is not sure which factors contribute the most to their posi- tion along the second axis. It is known that A. pyrenaicum is an alkaliphilous taxon mainly occurring at pH > 7 and a ni- trogen-autotrophic taxon, tolerating elevated concentrations of organically bound nitrogen (Van Dam et al. 1994, Noga et al. 2014). A. pyrenaicum had an optimum in oligo- to me- sotrophic calcium rich waters with medium to high concen- tration of electrolytes (Krammer and Lange-Bertalot 1991). It was very abundant in the territory of the Podkarpacie Province, Poland, in the upper course of the Wisłok River (Noga 2012). Fig. 3. Principal components analysis (PCA) of selected diatom indices (EPI-D – eutrophication and/or pollution index – diatom based, IPS – specifi c pollution index, IBD-biological diatom in- dex, CEE – commission for economical community metric, TDI – trophic diatom index) with sampling time (April, July and Sep- tember) and sampling site (ML1–ML5) as supplementary vari- ables. JAKOVLJEVIĆ O. S., POPOVIĆ S. S., VIDAKOVIĆ D. P., STOJANOVIĆ K. Z., KRIZMANIĆ J. Ž. 204 ACTA BOT. CROAT. 75 (2), 2016 Many taxa already mentioned above occurred at most of the sampling sites and were the dominant taxa in the diatom community. Achnanthidium minutissimum, Achnanthidium pyrenaicum and Amphora pediculus were dominant taxa in all three seasons. In general, in our study a great diversity of diatom species was documented and a similar diversity was found in an investigation of the Porsuk River in Turkey (Solak 2011) and the Raška River in Serbia (Vidaković 2013). Navicula, Gomphonema and Nitzschia are usually the most numerous genera in rivers in Europe, and in Serbia as well (Solak 2011, Andrejić 2012, Noga et al. 2013, Vidaković 2013, Vasiljević et al. 2014). As mentioned, the lowest number of diatom taxa was recorded at ML1. Results showed that the sampling site ML1 had the lowest values of ammonia and total phospho- rous and highest values of oxygen and water hardness (seen also in DCA ordination diagram), as expected, due to the fi shpond located directly after the ML1. The higher number of diatom taxa from the second locality was caused by the higher level of nutrients, especially phosphates. The water quality studies of the Mlava River during April, July and September 2011 indicated good and high water qualities. This was confi rmed by most of the mea- sured physico-chemical characteristics which indicated class I water quality. Diatom indices calculation showed that at the studied sites there are no major variations in wa- ter quality, regardless of the trout pond which is located be- tween the fi rst and second locality. Based on PCA analysis, a correlation was noticed between the selected indices, indi- cating their applicability to the rivers in Serbia. The EPI-D index was used in monitoring seven rivers located in the central eastern Apennine sector (Italy) (Torrisi and Dell’ Uomo 2006). As in our study, a high correlation was shown between this index and the IPS and IBD indices. IPS and CEE are the most sensitive to eutrophication and organic pollution (Descy and Coste 1991). CEE had the highest value at sampling site ML1, which can be explained by the position of the pond (between the fi rst and the second local- ity), although these values still are within the fi rst class quality. According to Szabó et al. (2004), among IPS, IBD, CEE and EPI-D diatom indices, which are good for Hun- garian rivers, IPS is probably the best. The suitability of the index is dependent on the percentage of taxa used for the index calculation. In our study, the average percentage of taxa used for the IPS calculation is 100, so IPS is the most suitable index for this reason. TDI is not correlated with the other four indices in our study, which is expected, since it is the only index in which the values range from 1 to 100 while the values of all other indices range from 1 to 20. Values of all fi ve indices increase in the second and third season which indicates the improvement of water quality. The data provided by this study can be used for the devel- opment of a biomonitoring tool for the rivers in Serbia. Acknowledgments Financial support was provided by the Ministry of Edu- cation and Science of the Republic of Serbia (Project No. TR 037009). References Ács, É., Borics, G., Fehér, G., Kiss, K.T., Reskóne, N.M., Stenger- Kovács, Cs., Vábríró, G., 2009: Implementation of the Euro- pean Water Framework Directive to assessment the water quality of Hungarian running waters with diatoms. Datomed- edelingen 33, 29–33. Àcs, É., Szabó, K., Tóth, B., Kiss, K.T., 2004: Investigation of benthic algal communities, especially diatoms of some Hun- garian streams in connection with reference conditions of the water framework directives. Acta Botanica Hungarica 46, 255–277. Andrejić, J., 2012: Floristic-ecological analysis of diatoms (Bacil- lariophyta) from the Nišava River and tributaries Jerma and Temska Rivers. Doctoral Dissertation, University of Belgrade, Belgrade. Babić Mladenović M., 2009: Sub-basin level fl ood action plan: Velika Morava River basin and right Danube tributaries be- tween the Sava River Mouth and RSBG Border. Ministry of Agriculture, Forestry and Water Management, Republic Di- rectorate for Water, in cooperation with Institute for Develop- ment of Water Resources »Jaroslav Černi«, Belgrade, and Re- public Hydrometeorological Service of Serbia, Belgrade and Ministry of Environment and Water, Sofi a Danube River Ba- sin Directorate, Sofi a. Chételator, J., Pick, J., Morin, A., Hamilton, G., 1999: Periphyton biomass and community compoition in rivers of different nu- trient status. Canadian Journal of Fisheries and Aquatic Sci- ences 56, 560–569. Coste in Cemagref, 1982: Etude des méthodes biologiques quanti- tative d’appréciation de la qualité des eaux. Rapport Division Qualité des Eaux Lyon – Agence fi nancière de Bassin Rhône- Méditerranée-Corse, Pierre-Bénite. Coste, M., Boutry, S., Tison-Rosebery, J., Delmas, F., 2009: Im- provements of the Biological Diatom Index (BDI): Descrip- tion and effi ciency of the new version (BDI-2006). Ecological Indicators 9, 621–650. Dell’Uomo, A., 2004: L’Indice Diatomico de Eutrofi zzacione/ Polluzione (EPI-D) nel Monitoraggio delle Acque Correnti. Linee Guida. APAT: Roma. Denić, LJ., Čađo, S., Đuraković, A., Novaković, B., Dopuđa- Glišić, T., Veljković, N., Stojanović, Z., Milovanović, J., Domanović, M., 2015: Status of surface waters of Serbia: Analysis and elements for monitoring design. Agencija za zaštitu životne sredine, Ministarstvo poljoprivrede i zaštite životne sredine, Republika Srbija (In Serbian). Descy, J. P., Coste, M., 1991: A test of methods for assessing wa- ter quality based on diatoms. Verhandlungen der Internationa- len Vereinigung für Theoretische und Angewandte Limnolo- gie 24, 2112–2116. Hofmann, G., Werum, M., Lange-Bertalot, H., 2013: Diatomeen im Süßwasser – Benthos von Mitteleuropa. Bestimmungsfl ora Kieselalgen für die ökologische Praxis. Koeltz Scientifi c Books, Königstein. Jakovljević, O., Krizmanić, J., Cvijan, M., 2014: Water quality as- sessment of the DTD hydrosystem by diatom indices. Matica Srpska Journal for Natural Sciences 127, 22–33. WATER QUALITY MONITORING USING BENTHIC DIATOMS ACTA BOT. CROAT. 75 (2), 2016 205 Kawecka, B., 2012: Diatom diversity in streams of the Tatra Na- tional Park (Poland) as indicator of environmental conditions. W. Szafer Institute of Botany, Polish Academy of Sciences, Kraków. Kelly, M. G., Whitton, B. A., 1995: The trophic diatom index: a new index for monitoring eutrophication in rivers. Journal of Applied Phycology 7, 433–444. Krammer, K., Lange-Bertalot, H., 1986: Bacillariophyceae. 1. Teil: Naviculaceae. In: Ettl, H., Gerloff, J., Heynig, H., Mol- lenhauer, D. (eds.), Süßwasserfl ora von Mitteleuropa 2/1, 1–876. Gustav Fischer Verlag, Jena. Krammer, K., Lange-Bertalot, H., 1991: Bacillariophyceae. 4. Teil: Achnanthaceae. Kritische Ergänzungen zu Navicula (Lineola- te) und Gomphonema. In: Ettl, H., Gartner, G., Gerloff, H. Hey- ning, Mollenhauer, D. (eds.), Süsswasserfl ora von Mitteleuro- pa 2/4, 1–437. Gustav Fischer Verlag, Stuttgart & New York. Krizmanić, J., Subakov Simić, G., Predojević, D., 2013: Algae as water quality bioindicators of the River Djetinja. Proceedings 6 International Conference „Water & Fish«, Belgrade-Zemun, 342–348. Kumar, N., Oommen, C., 2011: Phytoplankton composition in rela- tion to hydrochemical properties of tropical community wetland. Applied Ecology and Environmental Research 9, 279–292. Lecointe, C., Coste, M., Prygiel, J., 1993: „Omnidia«: software for taxonomy, calculation of diatom indices and inventories management. Hydrobiologia 269/270, 509–513. Manojlović, P., Mustavić, S., Mladenović, B., 2012: Chemical and mechanical water erosion ratio in the Mlava River basin. Bul- letin of the Serbian Geographical Society 92, 39–46. Noga, T., 2012: Diversity of diatom communities in the Wisłok River (SE Poland). In: Wołowski, K., Kaczmarska, I., Eh- rman, J.M., Wojtal, A.Z. (eds.), Phycological Reports: Current advances in algal taxonomy and its applications: phylogenet- ic, ecological and applied perspective, 109–128. Institute of Botany Polish Academy of Sciences, Krakow. Noga, T., Stanek-Tarkowska, J., Kochman, N., Peszek, Ł., Pajac- zek, A., and Woźniak, K., 2013: Application of diatoms to as- sess the quality of the waters of the Baryczka stream, left-side tributary of the River San. Journal of Ecological Engineering 14, 8–23. Noga, T., Stanek-Tarkowska, J., Pajączek, A., Kochman, N., Peszek, Ł., 2014: Ecological assessment of the San River wa- ter quality on the area of the San valley Landscape park. Jour- nal of Ecological Engineering 15, 12–22. Offi cial gazette of the Republic of Serbia, 2011: Ordinance on the parameters of the ecological and chemical status of surface waters and the parameters of the chemical and quantitative status of groundwater. Službeni glasnik RS, br. 74/2011 (in Ser bian). Pajączek, A., Musiałek, M., Pelczar, J., Noga, T., 2012: Diversity of diatoms in the Mleczka River, Morwawa River and Różanka Stream (tributaries of the Wisłok River, SE Poland), with par- ticular reference to threatened species. In: Wołowski, K., Kacz- marska, I., Ehrman, J.M., and Wojtal, A.Z. (eds.), Phycological reports: current advances in algal taxonomy and its applica- tions: phylogenetic, ecological and applied perspective, 129– 152. Institute of Botany Polish Academy of Sciences, Krakow. Poikane, S., Kelly, M., Cantonati, M., 2016: Benthic algal assess- ment of ecological status in European lakes and rivers: Chal- lenges and opportunities. Science of the Total Environment 568, 603-613.. Rimet, F., 2009: Benthic diatom assemblages and their correspon- dence with ecoregional classifi cations: a case study of rivers in north-eastern France. Hydrobiologia 636, 137–151. Rumeau, A., Coste, M., 1988: Initiation systematics of freshwater diatoms for practical use in a generic diatomic index. Bulletin Francais de la Peche et de la Pisciculture 309, 1–69. Rusanov, A., Stanislavskaya, E., Acs, E., 2009: Distribution of periphytic diatoms in the rivers of the Lake Ladoga basin (Northwestern Russia). Acta Botanica Croatica 68, 301–312. Solak, C. N., Àcs, É., 2011: Water quality monitoring in European and Turkish rivers using diatoms. Turkish Journal of Fisheries and Aquatic Sciences 11, 329–337. Solak, N.C., 2011: The application of diatom indices in the upper Porsuk Creek Kütahya – Turkey. Turkish Journal of Fisheries and Aquatic Sciences 11, 31–36. Stevenson, R. J., Bothwell, M. L., Lowe, R. L., 1996: Algal ecol- ogy: freshwater benthic ecosystems. Academic Press. Szabó, K., Kiss, K., Ector, L., Kecskés, M., Ács, E., 2004: Benthic diatom fl ora in a small Hungarian tributary of River Danube (Rákos-stream). Algological Studies 111, 79–94. Ter Braak, C.J.F., Šmilauer, P., 2012: Canoco reference manual and user‘s guide: software for ordination, version 5.0. Micro- computer Power, Ithaca, USA. Toman, M., Grošelj, A., Zelnik, I., 2014: The infl uence of selected factors on the distribution of epilithic diatoms in a torrential river the Kamniška Bistrica (Slovenia). Acta Botanica Croati- ca 73, 447–463. Tornés, E., Cambra, J., Gomà, J., Leira, M., Ortiz, R., Sabater, S., 2007: Indicator taxa of benthic diatom communities: a case study in Mediterranean streams. Annales de Limnologie-Inter- national Journal of Limnology 43, 1–11. Torrisi, M., Dell’Uomo, A., 2006: Biological monitoring of some Apennine rivers (central Italy) using the diatom-based eutro- phication/pollution index (EPI-D) compared to other Europe- an diatom indices. Diatom Research 21, 159–174. Van Dam, H., Mertens, A., Sinkeldam, J., 1994: A coded checklist and ecological indicator values of freshwater diatoms from the Netherlands. Netherlands Journal of Aquatic Ecology 28, 117–133. Vasiljević, B., Krizmanić, J., Ilić, M., Marković, V., Tomović, J., Zorić, K., Paunović, M., 2014: Water quality assessment based on diatom indices – small hilly streams case study. Wa- ter Research and Management 4, 31–35. Vidaković, D., 2013: Assessment of the ecological status of Raška River based on epilithic diatoms. Master Thessis, University of Belgrade, Belgrade. Virtanen, L., Soininen, J., 2012: The roles of environment and space in shaping stream diatom communities. European Jour- nal of Phycology 47, 160–168. WFD CIS Guidance Document No. 7., 2003: Monitoring under the WFD, Produced by Working Group 2.7-Monitoring, Euro- pean Communities. WFD, 2000: Water Framework Directive – Directive of European Parliament and of the Council 2000/60/EC – Establishing a Framework for Community Action in the Field of Water Poli- cy. European Union, the European Parliament and Council, Luxembourg. WFD-UKTAG, 2014: Phytobenthos – Diatoms for assessing river and lake ecological quality (River DARLEQ2). Water Frame- work Directive – United Kingdom Technical Advisory Group (WFD-UKTAG), Scotland.