PRELIMINARY CHECKLISTS FOR APPLYING SERCON ( SYSTEM FOR EVALUATING RIVERS FOR CONSERVATION ) TO RIVERS IN SERBIA

This paper describes the first steps in gathering biological data to assess the conservation value of rivers in Serbia, using SERCON (System for Evaluating Rivers for Conservation). SERCON was developed in the UK to improve consistency in assessments of river ‘quality’ by using a scoring system to evaluate habitat features and species groups, catchment characteristics, and the potential impacts to which river systems may be subjected. This paper provides checklists for aquatic, semiaquatic and marginal plants, macroinvertebrates, fish and birds associated with rivers in Serbia, collated from a wide range of published and unpublished sources. These lists should be regarded as provisional because few wide-ranging biological surveys have been carried out specifically on Serbian rivers; further revisions are likely as more information becomes available in future. Ultimately, the work will benefit regulators and decision-makers with responsibility for river management under the new Water Law, and contribute to river protection and conservation in Serbia.


INTRODUCTION
There are many reasons for assessing the nature conservation value of rivers, such as assisting with the production of catchment management plans, contributing to the process of environmental impact assessment, targeting appropriate rivers for rehabilitation, or selecting important areas for protection.One of the fundamental problems with this approach is the inherent subjectivity of many of the criteria.While it is recognized that subjectivity will always be an element of conservation evaluation, such a process becomes effectively unworkable if values are left undefined.Unless an evaluation procedure can be used that is rigorous and repeatable, it is also extremely difficult both to describe the value of rivers in anything other than vague terms, and to compare the relative merits of different rivers (Boon et al., 1997).
The broad concept of 'conservation value' may be understood in a variety of ways.Even though these are not mutually exclusive, the factors which contribute to 'value' in each case are often not close-ly related, so it is inappropriate to attempt to produce some total assessment of 'conservation quality' .SERCON (System for Evaluating Rivers for Conservation, Boon et al., 1997) is concerned only with nature conservation value and its assessment.This paper marks the start of a process for gathering data to apply SERCON to rivers in Serbia.

Background to SERCON
The prime motivation for developing a new technique for river conservation assessment was twofold: a perceived need to increase the breadth, rigor and repeatability of evaluations, and shifting the focus from merely seeking ways of protecting the best to managing, improving and restoring river resources across the full spectrum of conservation value (Boon et al., 2009).Moreover, with the demands of the Water Framework Directive, the need for new river monitoring programs, and a growing interest in river restoration, relevant tools for evaluating river 'quality' are very much in demand.
Work on SERCON (System for Evaluating Rivers for Conservation) began in the early 1990s, led by SNH in partnership with the statutory conservation and environment agencies in the UK (Boon et al., 1997).A re-design of the system to create SER-CON 2 (Boon et al., 2002) introduced several significant changes, although the general principles and structure remain the same -essentially a technique for scoring a wide range of habitat features and species groups, catchment characteristics, and the potential impacts to which river systems may be subjected.Information on each attribute is used to create a picture of a river in terms of accepted conservation criteria such as Naturalness, Representativeness, and Rarity.A separate part of the system records other background data (altitude, catchment land-use, phosphorus levels) helping to place each nature conservation assessment in a geographical, geological and water quality context.The list of the attributes and criteria evaluated in SERCON 2 is given in Boon et al. (2002).
The geographical focus in SERCON is the river channel, river banks, riparian zone, and the floodplain, but with the evaluation also taking account of broader catchment characteristics.The primary output from SERCON is a suite of conservation indices (from 0-100), one for each of the six main criteria, and an impact index, derived from information on each attribute.A seventh unscored category (Additional Features of Importance) allows the user to draw attention to unique or unusual features of the river.Each evaluation is restricted to a river reach known as an 'ECS' -an abbreviation in SERCON 2 for 'Evaluated Corridor Section' .

Practical applications
One of the principal objectives of SERCON was to widen the scope of river conservation evaluation from a narrow focus (on aquatic plants or rare species, for example) to the full spectrum of conservation value.However, there are occasions where scientific data are limited when it may not be possible, or even desirable, to allocate resources for a full SERCON evaluation In response to the need for a 'slimmer' version of SERCON, a series of 'SERCON Applications' was developed, each allowing flexibility to 'pick and mix' SERCON attributes for specific purposes.
Seven applications have been designed: • Reporting Special Natural Characteristics • Assessment of Aspects of the Flora or Fauna

• Monitoring
This helps to make SERCON more relevant to particular user groups, such as conservation agencies or environmental consultants.
To operate SERCON, the user needs to obtain and collate a range of physical, chemical and biological data.Standard checklists of plants and animals associated with rivers are an essential part of SERCON in order to ensure consistency.However, because SERCON was originally developed for use in the UK, those carrying out evaluations elsewhere need to refer to checklists specific to their own countries, or preferably to individual ecoregions.Hence, the aim of this study was to provide checklists for aquatic, semi-aquatic and marginal plants, aquatic macroinvertebrates, fish and birds, associated with rivers in Serbia that can be used in the recently developed SERCON software at the Faculty of Science, University of Novi Sad.These checklists should be regarded as provisional and subject to further refinement.In general, few surveys have been carried out on the biota of Serbian rivers, apart from those that collect data on freshwater invertebrates for routine water quality monitoring.It is intended that future work, specific to running waters, will help in revising the lists -for example, by adding species that were hitherto unrecorded, or by removing species subsequently found to be associated mainly with other types of habitat rather than rivers.

Aquatic and marginal plant checklist
There is a growing need to catalogue riverine, ditch, lake and pond organisms, to assess their wildlife value and to understand the impact of management on their conservation in the face of advancing threats such as water pollution and river engineering.For rivers, this usually involves recording plants along transects, while for standing waters it is usually combined with walkover and grapnel sampling (Rodwell, 1995), where aquatic, emergent and bankside plants may be sampled for different purposes.As aquatic plants are generally considered to be sensitive to physical alteration in streams, being in close contact with the environmental conditions of rivers through the root-system, they are frequently used to establish the ecological quality of aquatic ecosystems (O'Hare et al., 2010;Radulović et al., 2010).
However, one of the main problems when recording freshwater plants is how to define them.Preston and Croft (1997) pointed out that the boundaries of fresh waters are blurred and that many plant species are well adapted to boundary zones liable to fluctuation in water level.They recognized as freshwater aquatic plants those species they regarded as characteristically growing with at least their basal parts permanently in water throughout the year.This criterion, however, still leaves space for misinterpretation.As such, the difficulties in defining aquatic and semi-aquatic plants have led to different recorders adopting differing views on what constitutes 'aquatic plants' .This emphasizes the need to use a standard, published checklist, the absence of which has been a major drawback of most freshwater vegetation surveys in the past.
The list corresponds with the National EUNIS classification (Lakušić et al., 2005), containing all aquatic macrophytes as well as all semi-aquatic and marginal species, including their conservation status according to Stevanović (1999), IUCN (2011) and SGRS (2010a).
The wide diversity of the freshwater macroinvertebrate fauna poses a serious challenge in establishing standard lists of species, especially if there are no national checklists or they need to be updated.Owing to the scarcity of information on the benthic macroinvertebrate fauna in Serbia, family level rather than species level has been used in environmental assessment, despite the potential limitations of taxonomic resolution.In SERCON, both familyand species-level data may be used in conservation assessment.Moreover, Wright et al. (1998a) demonstrated that both family richness and BMWP family richness are good surrogates for macroinvertebrate species richness.
It is important to recognize that the lists in Tables 2-4 represent merely the first attempt to collate most of the families in the Serbian freshwater invertebrate fauna for the purposes of assessing the conservation status of rivers and streams.Assessments of invertebrate naturalness in SERCON are made using the River Invertebrate Prediction System (RIVPACS) (Armitage et al., 1983;Clarke et al., 1994).RIVPACS provides site-specific predictions of the macroinvertebrate fauna expected based on environmental features in the absence of major environmental stress (Wright et al., 1997;Wright et al., 1998b).In the absence of a RIVPACS-type system in Serbia, it would be necessary to define the composition of the macroinvertebrate community at reference condition for each river or river type.

The Fish checklist
The list of species (Table 5) of finned fish and lampreys was compiled for rivers in Serbia, within the catchment areas of the Black, Adriatic and Aegean seas.The list was created using data given in Simonović and Nikolić (1996) and Simonović (2001), supplemented with records on new discoveries of lamprey and fish species reported for the territory of the Republic of Serbia (Hegediš et al. 1991;Sekulić et al. 1998Sekulić et al. , 1999;;Simonović and Nikolić 1996;Simonović 1999;Simonović and Marić 2006;Simonović et al. 1998Simonović et al. , 2006aSimonović et al. , 2006bSimonović et al. , 2007Simonović et al. , 2010aSimonović et al. and 2010b;;Cakić et al. 2000Cakić et al. , 2004;;Cvijanović et al. 2005;Šipoš et al. 2004;Lenhardt et al. 2006Lenhardt et al. , 2011;;Marić et al. 2004Marić et al. , 2006Marić et al. , 2011 ) ).The nomenclature of fish and lamprey species relies on the phylogenetic species concept (Kottelat and Freihof, 2007) supplanting the previous nomenclature which followed the biological species concept, with sub-species specific for particular river drainages.The largest rate of sub-endemism occurs in the Adriatic Sea catchment (Beli Drim and Plav Rivers) in the southwestern part of Serbia, followed by those in the Aegean Sea (Lepenac and Dragovištica Rivers) and the Black Sea catchments (the Danube River).
The list encompasses both native and non-native species, with translocations of native species occur-
In addition to the number of fish species, their abundance indicates that the Sava River is the Ser-  bian inland water most threatened by the impact of non-native fish species.Many reservoirs also suffer from invasions, whose extent is related to the age and accessibility of the water body, as shown by the Vlasina reservoir (Simonović andNikolić 1996, 1997).
In some cases the occurrence of particular listed fish species is uncertain (regardless of whether they are indigenous or not).This may result from unreliable records (as for Pungitius platygaster and Coregonus peled), acknowledged misidentifications (Ictalurus nebulosus), lack of data about their status in both inaccessible areas (for Salmo marmoratus in southwestern Serbia) and in large catchment systems (Acipenser gueldenstaedti and Salmo labrax), the current status of feral species, or discrepancy between the recent classification and former taxonomic assignment (Salmo labrax).Nevertheless, recent records of Acipenser nudiventris and Acipenser gueldenstaedti suggest that the available data on fish and lamprey diversity should be maintained until records can eventually be verified.

The Bird checklist
In total, 351 species of birds have been recorded in Serbia (Supplement 1, Table 6 and Table 7) (Puzović and Grubač 2000;Puzović (Ed.) 2000;Puzović, 2008;Puzović et al., 2009, Tucakov et al., 2009), of which 45.2% (159) occur in wetlands or open waters and 38 are particularly associated with rivers.Some of these waterfowl are present in high density: 9800-       is an important factor in their protection.Competent Water Authorities, together with authorities for human health, environmental protection and nature conservation will designate protected areas.These bodies will also establish specific monitoring programs (or modify regular ones), if and where necessary, in accordance with specific requirements and other relevant legislation.The newly adopted water status classification system, as well as the monitoring requirements under the Water Law, have yet to be implemented.Thus, the publication of these locally/regionally adjusted checklists to accompany SERCON as a holistic nature conservation tool is timely.Fur-ther development will benefit regulators, competent authorities and all stakeholders and decision-makers in water protection and conservation.Of course, as Boon and Freeman (2009) point out, the challenge in river conservation is to find ways of moving from survey and inventory into evaluation, from evaluation into conservation planning, and from conservation planning into practical conservation management.In Serbia, as in many other countries, there is still a long way to go.

Table 2
Checklist of freshwater macroinvertebrate families, grouped in accordance with the original list of BMWP families given in SERCON Manual(Boon et al., unpublished)and added families of Oligochaeta.

Table 5 .
The Fish check-list

Table 7 .
Breeding riverine birdsSystem for Long-Term Research and Integrated Management of Ecosystems' (Biosensing tehnologije i globalni sistem za kontinuirana istraživanja i integrisano upravljanje ekosistema) III 043002 grant, while the biodiversity dataset was prepared the project Plant biodiversity of Serbia and the Balkans -assessment, sustainable use and protection (Biodiverzitet biljnog sveta Srbije i Balkanskog poluostrva -procena, održivo korišćenje i zaštita) 173030 Grant, supported by Ministry of Education and Science, Republic of Serbia