**Priority substances**

Prespa Lake waters were also found to contain 20 priority substances out of more than 70 analysed during March and July, 2010. The same applied for the river water bodies, with *Bis(2-Ethylhexyl)phthalate* and *Dibutilphthalate* dominating in the samples of Prespa Lake waters. But, there was also a marked presence of *Benzo (a) pyrene, Benzo (a) anthracene* and *Naphthalene* (Fig. 26).

Considering the presence of pesticides or their residues in the Prespa Lake waters (Fig.27), *Gama HCH (or Lindan)* is dominantly found on all sampling sites followed by *Heptachlor* which

Contrary to presented results for the river water bodies (Fig. 13), the concentrations of heavy metals detected in the waters of selected Prespa Lake sampling sites (Fig.25) point to *copper, iron* and *zinc* as dominant metals. For *copper* the detected values were almost entirely in III-IV category, but the increased presence of the other two metals in detected concentrations

There is also a marked presence of the toxic *mercury* and *arsenic* in Prespa Lake waters. Their increased detected concentrations in July 2010 clearly support the argument for intensified human activities in the spring period; *mercury* appears at the L2 sampling site as a result of the Golema River influence and is detected almost in all samples with concentrations high above the V water quality class. Its presence is striking on L5 (v.Dolno Dupeni) with almost eight times increased concentration compared to the 1 mg\*L-1 level for V quality class. *Arsenic* is also present in all sampled waters of Prespa Lake, but in much lower concentrations than *mercury*. It rose to III-IV water quality range only in L4 (v.Nakolec – mouth waters of river Brajčinska) in July 2010. Nevertheless, its accumulation and

Heavy metals in Prespa Lake

As (30-50 μg/l) III-IV class

Hg (>1 μg/l) V class Cu (>10-50 μg/l) III-IV class

Fig. 25. Detected concentrations of heavy metals in the water of selected sampling sites of

0

1

5

10

15

20

As [μg/L] Al [μg/L] Cd [μg/L] Cr [μg/L] Fe [μg/L] Mn [μg/L] Ni [μg/L] Pb [μg/L] Zn [μg/L] Cu [μg/L] Hg [μg/L]

25

Cu, Hg

March April July March April July March April July March April July March April July L1 L2 L3 L4 L5

Prespa Lake waters were also found to contain 20 priority substances out of more than 70 analysed during March and July, 2010. The same applied for the river water bodies, with *Bis(2-Ethylhexyl)phthalate* and *Dibutilphthalate* dominating in the samples of Prespa Lake waters. But, there was also a marked presence of *Benzo (a) pyrene, Benzo (a) anthracene* and

Considering the presence of pesticides or their residues in the Prespa Lake waters (Fig.27), *Gama HCH (or Lindan)* is dominantly found on all sampling sites followed by *Heptachlor* which

**Heavy metals** 

Prespa Lake.

0

20

40

60

80

100

120

140

**Priority substances** 

*Naphthalene* (Fig. 26).

also confirms the prolonged input.

persistence in Prespa Lake waters is evident.

is usually in concentrations high above the permissible 1 ng\*L-1. It is interesting to notice that the L2 sampling site in the vicinity of the mouth waters of Golema River was found with the lowest number of detected priority substances, contrary to the expected and detected pressures coming from this water body. The other sampling sites along the North-East coast of the lake (L3-L5) and the deepest part on L1 sampling site had a much higher number of detected priority substances and maximal values of separate chemicals. These findings corroborate the proposed intensive mixing of the Prespa Lake waters with significant underwater currents that are spreading the pollution impact to a much wider area.

Fig. 26. Priority substances (in ppm) detected in waters of selected Prespa Lake sampling sites. Priority substances detected in Prespa Lake pose a significant hazard to biota and humans.

Fig. 27. Priority substances (in ppb) detected in the waters of selected Prespa Lake sampling sites.

Environmental Changes in Lakes Catchments

their patterns and role in the ecosystem.

sampling sites.

0

0.5

1

1.5

2

2.5

3

μ**g/kg**

**Ecological quality elements** 

*phytobenthos, invertebrate fauna, macrophytes* and *fish.* 

6 WFD-Guidance document No.7: Monitoring under the WFD, 2003.

as a Trigger for Rapid Eutrophication – A Prespa Lake Case Study 103

The final analysis of the priority substances in Prespa Lake watershed was performed on sediment samples from the selected sampling sites in the lake (Fig. 30). At this point, only *gamma-HCH (Lindan)* was detected in the sediments of all sampling sites while the results for the L4 sampling site revealed sedimentation of the greatest number of analysed substances. Further monitoring of priority substances in Prespa Lake watershed should include far more frequent samplings and other media (like biota) in order to obtain overall conclusions about

**PRIORITY SUBSTANCES IN PRESPA LAKE SEDIMENTS**

gama-HCH beta-HCH PCB HCB

alfa-Endosulfan 4,4-DDD

Fig. 30. Results on priority substances obtained by analysis of sediments in Prespa Lake

L1 L2 L3 L4 L5

Investigations of biological quality elements in selected Prespa Lake sampling sites (L1-L5) were performed on all selected WFD's Ecological Quality Elements6 - *phytoplankton,* 

Benthic diatom communities in Prespa Lake have been recently well documented (Levkov et al., 2006) and again confirmed with the performed investigations. Nevertheless, their ecological preferences are more elusive since very limited investigations in that context have been performed so far on the lake. The greatest difference in diatom composition between Eastern and Western coast of Prespa Lake was observed in the littoral zone. The bottom of the West coast (from Stenje village to Perovo village) is covered with organic sediment. Beside dominant species *Cavinula scutelloides, Navicula rotunda, N. subrotundata* and *Amphora pediculus*, characteristic species from genera *Aneumastus* and *Sellaphora* are frequent in the benthic communities. The Eastern coast (D. Dupeni village to Pretor) is mainly covered by sand or a mixture of sand and organic sediment. In this region, several *Navicula* sensu stricto taxa are sub-dominant. It is supposed that distribution of these species is influenced by substrate. A few species so far known only from Lake Prespa could be found on both sides.

By comparing the obtained results on priority substances for the river water bodies and Prespa Lake sampling sites (Fig. 28 and 29) interesting correlations could be formulated. Substances detected in high concentrations in the rivers, like *Bis(2-Ethylhexyl)phthalate* or *gamma-HCH (Lindan)* remain high in the lake's waters as well. Although others that were not recorded in very high concentrations in rivers, such as *Dibutilphthalate* or *Heptachlor*, show much higher concentrations in the lake, while *PCB's* tend to disappear from the lake's waters. These findings shed light on the very complicated and unpredictable pathways the detected priority substances have in the Prespa Lake ecosystem and point to the fundamental necessity to monitor and reveal their final destiny and impact they pose to the ecosystem, biota and human health.

Fig. 28. The influence of river ecosystems on Prespa Lake regarding priority substances (in ppm).

Fig. 29. The influence of river ecosystems on Prespa Lake regarding priority substances (in ppb).

By comparing the obtained results on priority substances for the river water bodies and Prespa Lake sampling sites (Fig. 28 and 29) interesting correlations could be formulated. Substances detected in high concentrations in the rivers, like *Bis(2-Ethylhexyl)phthalate* or *gamma-HCH (Lindan)* remain high in the lake's waters as well. Although others that were not recorded in very high concentrations in rivers, such as *Dibutilphthalate* or *Heptachlor*, show much higher concentrations in the lake, while *PCB's* tend to disappear from the lake's waters. These findings shed light on the very complicated and unpredictable pathways the detected priority substances have in the Prespa Lake ecosystem and point to the fundamental necessity to monitor and reveal their final destiny and impact they pose to the

**Influence of the river water bodies on Prespa Lake priority substances concentrations in ppm**

Fig. 28. The influence of river ecosystems on Prespa Lake regarding priority substances (in

1 Istočka 2 Golema 1 Golema 2 Golema 3 Golema 4

March July March July March July March July March July March March March July July Julay March July July March March March July March

Anthracene Benzo (a )anthracene Benzo (a) pyrene Fluorene Alachlor Atrazine Naphthalene Dibutilphthalate Bis(2-Ethylhexyl)phthalate

**Influence of the river water bodies on Prespa Lake priority substances concentrations in ppb**

PCB (III-IV class) 1-10 ng/l Heptacholr andDDT (V class) > 1 ng/l

Golema 6 KurbinskaKranska

Golema 6 Kurbinska Kranska

1 Kranska 2

1 Kranska 2

Brajčinska 1 Brajčinska 2

Brajčinska 1 Brajčinska 2

Fig. 29. The influence of river ecosystems on Prespa Lake regarding priority substances (in

PCB-52 PCB-101 PCB-156 PCB-180 trans-Chlordane 2,4'-DDD 4,4'-DDD 2,4'-DDE 4,4'-DDE 4,4'-DDT alpha-Endosulfan alpha-HCH beta-HCH gamma-HCH (Lindan) delta-HCH Heptachlor cis-Heptachlor epoxide

March July March July March July March July March July March March March July July Julay March July July March March March July March

1 Golema 2 Golema 3 Golema 4

ecosystem, biota and human health.

L1 L2 L3 L4 L5 Istočka

L1 L2 L3 L4 L5 Istočka 1 Istočka 2 Golema

ppm).

ng/l

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 μg/l

ppb).

The final analysis of the priority substances in Prespa Lake watershed was performed on sediment samples from the selected sampling sites in the lake (Fig. 30). At this point, only *gamma-HCH (Lindan)* was detected in the sediments of all sampling sites while the results for the L4 sampling site revealed sedimentation of the greatest number of analysed substances. Further monitoring of priority substances in Prespa Lake watershed should include far more frequent samplings and other media (like biota) in order to obtain overall conclusions about their patterns and role in the ecosystem.

Fig. 30. Results on priority substances obtained by analysis of sediments in Prespa Lake sampling sites.

## **Ecological quality elements**

Investigations of biological quality elements in selected Prespa Lake sampling sites (L1-L5) were performed on all selected WFD's Ecological Quality Elements6 - *phytoplankton, phytobenthos, invertebrate fauna, macrophytes* and *fish.* 

Benthic diatom communities in Prespa Lake have been recently well documented (Levkov et al., 2006) and again confirmed with the performed investigations. Nevertheless, their ecological preferences are more elusive since very limited investigations in that context have been performed so far on the lake. The greatest difference in diatom composition between Eastern and Western coast of Prespa Lake was observed in the littoral zone. The bottom of the West coast (from Stenje village to Perovo village) is covered with organic sediment. Beside dominant species *Cavinula scutelloides, Navicula rotunda, N. subrotundata* and *Amphora pediculus*, characteristic species from genera *Aneumastus* and *Sellaphora* are frequent in the benthic communities. The Eastern coast (D. Dupeni village to Pretor) is mainly covered by sand or a mixture of sand and organic sediment. In this region, several *Navicula* sensu stricto taxa are sub-dominant. It is supposed that distribution of these species is influenced by substrate. A few species so far known only from Lake Prespa could be found on both sides.

<sup>6</sup> WFD-Guidance document No.7: Monitoring under the WFD, 2003.

Environmental Changes in Lakes Catchments

classification is given in Table 8.

diverse macrophyte vegetation.

**Benthic Quality Index – BQI (Profundal fauna)** 

as a Trigger for Rapid Eutrophication – A Prespa Lake Case Study 105

In contrast to the extensive use of benthic invertebrates in river monitoring, ecological assessment in lakes has instead focused mainly on the response of open-water phytoplankton (usually measured as concentrations of chlorophyll *a*) to nutrient (mainly phosphorus) enrichment (OECD, 1982) and, to a lesser extent, that of profundal or sublittoral communities (Dinsmore et al., 1993). While the WFD has proposed a need for monitoring of littoral communities in lakes, their use in regional monitoring of lakes in Europe has been very limited. The lack of incorporation of littoral invertebrates into lake monitoring programmes reflects a traditional and common view that the structural heterogeneity of lake littoral areas, and associated variable distribution of benthic macroinvertebrates, negates the feasibility of their use in ecological assessment (Downes et al., 1993). Many anthropogenic impacts that affect rivers (Boon, 1992) however, also affect lakes, and would be expected to drive changes in the littoral macroinvertebrate community. In the frame of this project, according to WFD requirements, categorisation of the Macro Prespa Lake based on macrozoobenthos was done. Macroinvertebrates from five sampling sites and different depth regions were collected. Detailed analyses on composition, abundance, diversity of benthic invertebrate fauna and relative contribution of sensitive and tolerant invertebrate taxa were performed. Based on the recommendation by the Swedish Environmental Protection Agency, two indices of bottom fauna were used for ecological assessment of Prespa Lake: the Benthic Quality Index (BQI) based on profundal fauna and the Shannon Diversity index (H') based on littoral fauna. The corresponding water

**Shannon Diversity Index – H'** 

>4 >3.00 high 3-4 2.33-3.00 good 2-3 1.65-2.33 moderate 1-2 0.97-1.65 poor <1 <0.97 bad

In Lake Prespa, the macrophyte vegetation shows relatively high species diversity in different parts of the littoral region. A high number of species is recorded at localities Golema River – 24, Asamati - 23 and v. Stenje – 21, while macrophyte species number is quite lower at Brajcino - 13 and Dolno Dupeni - 12. Recorded differences in the number of macrophyte species are most probably a result of different ecological conditions present at investigated localities, especially regarding nutrients. Namely, the presence of a higher species number at localities Golema River, Asamati and v.Stenje implies a very intensive anthropogenic influence. These areas of the littoral region have an increased presence of organic and inorganic material, what enables intensive growth and development of more

Due to a decrease of the Prespa Lake water level in the last decade, marsh vegetation (reed and other emerged plants) progressively expanded around the lake (previously submerged littoral). Obtained results show that the dominant emerged plant in all investigated localities was *Phragmites australis* (with a density of 5 according to five-point scale). The reed forms a

Table 8. Water classification based on BQI and H' for bottom fauna.

**(Littoral fauna) Water quality** 

Diatom assemblages on larger depths are dominated by *C. ocellata* complex, as well as species with heavily silicified valves as *Diploneis mauleri*, *Campylodiscus noricus*, *Cymatopleura elliptica* and *Navicula hasta*.

Phytoplankton in Prespa Lake is much more uniform. It is usually composed of planktonic diatoms, like the *Cyclotella ocellata* complex, with very rare presence of algae belonging to other taxonomic groups, like the chrysophyte *Dinobryon bavaricum* during the winter months (November-April). But, there is a rapid change in the dominance during the summer months when strong development (more than 90% of dominance) of potentially toxic cyanobacteria like *Anabaena* sp. or *Aphanizomenon* sp. are usually observed, and clearly documented during the latest investigations in the frame of this project (Fig. 31).

Fig. 31. Dominant algae in Prespa Lake. [A] Benthic diatoms, [B] Dominant blue-green cyanobacteria, chrysophyte *Dinobryon bavaricum* colony and a diatom *Cyclotella ocellata* in plankton.

Diatom assemblages on larger depths are dominated by *C. ocellata* complex, as well as species with heavily silicified valves as *Diploneis mauleri*, *Campylodiscus noricus*, *Cymatopleura* 

Phytoplankton in Prespa Lake is much more uniform. It is usually composed of planktonic diatoms, like the *Cyclotella ocellata* complex, with very rare presence of algae belonging to other taxonomic groups, like the chrysophyte *Dinobryon bavaricum* during the winter months (November-April). But, there is a rapid change in the dominance during the summer months when strong development (more than 90% of dominance) of potentially toxic cyanobacteria like *Anabaena* sp. or *Aphanizomenon* sp. are usually observed, and clearly documented during the latest investigations in the frame of this

[A] [B]

Fig. 31. Dominant algae in Prespa Lake. [A] Benthic diatoms, [B] Dominant blue-green cyanobacteria, chrysophyte *Dinobryon bavaricum* colony and a diatom *Cyclotella ocellata* in

*elliptica* and *Navicula hasta*.

project (Fig. 31).

plankton.

In contrast to the extensive use of benthic invertebrates in river monitoring, ecological assessment in lakes has instead focused mainly on the response of open-water phytoplankton (usually measured as concentrations of chlorophyll *a*) to nutrient (mainly phosphorus) enrichment (OECD, 1982) and, to a lesser extent, that of profundal or sublittoral communities (Dinsmore et al., 1993). While the WFD has proposed a need for monitoring of littoral communities in lakes, their use in regional monitoring of lakes in Europe has been very limited. The lack of incorporation of littoral invertebrates into lake monitoring programmes reflects a traditional and common view that the structural heterogeneity of lake littoral areas, and associated variable distribution of benthic macroinvertebrates, negates the feasibility of their use in ecological assessment (Downes et al., 1993). Many anthropogenic impacts that affect rivers (Boon, 1992) however, also affect lakes, and would be expected to drive changes in the littoral macroinvertebrate community.

In the frame of this project, according to WFD requirements, categorisation of the Macro Prespa Lake based on macrozoobenthos was done. Macroinvertebrates from five sampling sites and different depth regions were collected. Detailed analyses on composition, abundance, diversity of benthic invertebrate fauna and relative contribution of sensitive and tolerant invertebrate taxa were performed. Based on the recommendation by the Swedish Environmental Protection Agency, two indices of bottom fauna were used for ecological assessment of Prespa Lake: the Benthic Quality Index (BQI) based on profundal fauna and the Shannon Diversity index (H') based on littoral fauna. The corresponding water classification is given in Table 8.


Table 8. Water classification based on BQI and H' for bottom fauna.

In Lake Prespa, the macrophyte vegetation shows relatively high species diversity in different parts of the littoral region. A high number of species is recorded at localities Golema River – 24, Asamati - 23 and v. Stenje – 21, while macrophyte species number is quite lower at Brajcino - 13 and Dolno Dupeni - 12. Recorded differences in the number of macrophyte species are most probably a result of different ecological conditions present at investigated localities, especially regarding nutrients. Namely, the presence of a higher species number at localities Golema River, Asamati and v.Stenje implies a very intensive anthropogenic influence. These areas of the littoral region have an increased presence of organic and inorganic material, what enables intensive growth and development of more diverse macrophyte vegetation.

Due to a decrease of the Prespa Lake water level in the last decade, marsh vegetation (reed and other emerged plants) progressively expanded around the lake (previously submerged littoral). Obtained results show that the dominant emerged plant in all investigated localities was *Phragmites australis* (with a density of 5 according to five-point scale). The reed forms a

Environmental Changes in Lakes Catchments

*Toxic heavy metals and priority substances* 

*Conductivity (*

*(*μ*g\*L-1)*  μ

as a Trigger for Rapid Eutrophication – A Prespa Lake Case Study 107

*Reference conditions for the rivers in Prespa Lake watershed* 

<0.001

*circulare var. constricta, Diatoma hyemalis, Diatoma mesodon, Eunotia spp., Staurosirella pinnata, Hannea arcus, Psammothidium daonense, Amphipleura pellucida, Decussata hexagona, Luticola nivalis, Diadesmis perpusila, Krsticiella ohridana, Pinnularia* 

Red algae: *Lemanea fluviatilis.*

*alpinus, Baetis fuscatus, Baetis vernus, Potamophylax latipennis, Capnia vidua, Brachyptera risi, Nemoura cinerea, Austropotamobius torrentium, Astacus* 

*S\*cm-1)* <50

*Dominant algae* Diatoms: *Meridion circulare, Meridion* 

*Dominant benthic invertebrates Heptagenia sulphurea, Baetis rhodani, Baetis* 

Establishing the reference conditions for Prespa Lake (or any other lake) is a far more difficult task to perform. If the only reasonable and justified principle regarding every water ecosystem as a separate entity (the state-changed approach opposite to spatial state classification – Moss et al., 1997) is applied, than Prespa Lake cannot be compared for its reference parameters to any other lake (even with Lake Ohrid to which Prespa Lake is the major water source). This is even more important if the very turbulent and variable past of Prespa Lake is taken into account. Namely, the lake was formed by three rivers (underwater flows which are still detectable in the lake) which were constrained by karstic masses blocking their way to Ohrid Lake. Only then did the Prespa Lake ecosystem start to develop with a very variable surface area and volume in the past; there are numerous human constructions (buildings, roads) recorded at the lake's bottom today. All of these characteristics describe Prespa Lake as a very large water source body, intensively mixed by numerous sub-lacustrine sources of water and with very unstable water mass basically depending on climate, hydrologic regime and human activities. It is also a system in which

DSFI index – invertebrates ≥7

Table 9. Reference conditions for rivers in Prespa lake watershed.

**6.2 Reference conditions for Prespa Lake** 

*sudetica.*

*astacus* 

*Parameter (units)* Value *Dissolved oxygen (mg\*L-1)* >9

*pH* 6-7 *NHx-N ( mg\*L-1)* <0.05 *NOx-N ( mg\*L-1)* <0.6 *Total N ( mg\*L- )* <1.0 *PO4-P ( mg\*L-1)* <0.020 *Total P ( mg\*L-1)* <0.030

natural discontinuous belt around the lake comprised of numerous dense complexes. Other representatives of emergent vegetation were present in the belt of reed and in particular localities, where they formed almost pure associations. *Phalaris arundinacea*, *Typha latifolia*, *Typha angustifilia* were present with a density of 2, while *Shoenoplectus lacustris*, *Scirpus sylvaticus*, *Heleocharis pallustris* and *Cyperus longus* were present with a density of 1. Nevertheless, to obtain detailed information about changes in composition and spatial disposition of the vegetation, aquatic and marsh vegetation, long-term investigations are needed.

The fish population of the Prespa Lake is composed of 23 species of which 11 are autochthonous: *Alburnoides prespensis, Alburnus belvica, Anguilla anguilla, Barbus prespensis, Chondrostoma prespense, Cobitis meridionalis, Cyprinis carpio, Pelasgus prespensis, Rutilus prespensis, Salmo peristericus* and *Squalius prespensis*.

In the previous period 12 allochthonous species were introduced in Lake Prespa: *Carassius gibelio, Ctenopharyngodon idella, Gambusia holbrooki, Hypophthalmichthys molitrix, Lepomis gibbosus, Oncorhynchus mykiss, Parabramis pekinensis, Pseudorasbora parva, Rhodeus amarus, Salmo letnica, Silurus glanis* and *Tinca tinca*.

*Alburnus belvica* and *Rutilus prespensis* were caught in the greatest number of specimens at all investigated localities from Lake Prespa. On the contrary, *Anguilla anguilla*, *Chondrostoma prespense*, *Cobitis meridionalis*, *Pelasgus prespensis* and *Salmo peristericus* were not caught at any of investigated localities. Also, in the catches from all investigated localities *Cyprinis carpio* was present, but in very low numbers.

Populations of some of the introduced species are so reduced that they are very rarely present in representative and experimental fishing, like *Ctenopharyngodon idella*, *Hypophthalmichthys molitrix*, *Oncorhynchus mykiss*, *Parabramis pekinensis*, *Salmo letnica*, *Silurus glanis* and *Tinca tinca*. Others like *Carassius gibelio*, *Gambusia holbrooki*, *Lepomis gibbosus*, *Pseudorasbora parva* and *Rhodeus amarus* are present in the catches but in very low numbers.
