The status of freshwater pearl mussel in the Czech Republic: Several successfully rejuvenated populations but the absence of natural reproduction Academic research paper on "Biological sciences"

Accepted Manuscript

Title: The status of freshwater pearl mussel in the Czech Republic: several successfully rejuvenated populations but the absence of natural reproduction

Author: Ondrej P. Simon Ivana Vanickova Michal Bily Karel Douda Hana Patzenhauerova Jaroslav Hruska Alena Peltanova

PII: DOI:

Reference:

S0075-9511(14)00077-2

http://dx.doi.Org/doi:10.1016/j.limno.2014.11.004 LIMNO 25429

To appear in:

Received date: Revised date: Accepted date:

31-3-2014

17-11-2014

18-11-2014

Please cite this article as: <doi>http://dx.doi.org/10.1016/j.limno.2014.11.004</doi>

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

The status of freshwater pearl mussel in the Czech Republic: several successfully rejuvenated populations but the absence of natural reproduction

Ondrej P. Simon1,2,8, Ivana Vanickova3'4'8, Michal Bilv2, Karel Douda5, Hana Patzenhauerova6, Jaroslav Hruska7, Alena Peltanova

1T.G. Masaryk Water Research Institute, Podbabska 30, CZ-16000 Prague 6, Czech Republic 2Faculty of Environmental Sciences, Czech University of Life Sciences Prague, Kamycka 129, CZ-16521 Prague 6, Czech Republic

3 Nature Conservation Agency (NCA CR), Kaplanova 1931/1, CZ-14800, Prague, Czech

Republic

4 Biology Centre of the Academy of Sciences of the Czech Republic, Institute of Hydrobiology,

Na Sadkach 7, CZ-37005 Ceske Budejovice, Czech Republic 5Faculty of Agrobiology, Food and Natural Resources, Czech University of Life Science,

Kamycka 129, CZ-16521 Prague 6, Czech Republic 6Institute of Vertebrate Biology, Academy of Sciences of the Czech Republic, Kvetna 8, CZ-603

65 Brno, Czech Republic 7Miletinky, CZ-383 01 Prachatice, Czech Republic 8These authors contributed equally to this work.

Keywords: Margaritifera margaritifera, action plan, Unionidae, conservation biology, long-term study, rejuvenation

Author for correspondence: simon@vuv.cz Abstract

The freshwater pearl mussel was historically abundant in many streams and rivers in the Elbe, Oder and Danube Basins in the Czech Republic, Central Europe. By the 21st century, the mussels had become extinct in the lower and middle altitudes, and current populations are only present near the upper limit of their natural range. The current population of this mussel is estimated to be only 1% of the historical abundance. The population decline was related to the negative impacts of pollution from industry, intense agriculture, forestry and sewage water. The freshwater pearl mussel habitat has also been impacted by watercourse regulations and has been fragmented by dams and weirs. All of these impacts have resulted in failure of the reproductive cycle; the last significant cohort of juveniles settled approximately 30-40 years ago. Therefore, this species is considered critically endangered, and an action plan was developed to conserve the populations in the Czech Republic. Special measures were conducted between 1984 and 2005 to improve the age structure of elderly populations. Fish infected with millions of glochidia were released in two locations, and over 50,000 captive-bred juveniles that were three to five years old were released in 7 locations. Only the latter approach resulted in a small number of subadults that gradually emerged from the substratum to the bottom surface, as confirmed by monitoring efforts. Despite simultaneous efforts to restore mussel habitat over the last 25 years, natural reproduction still does not occur in the Czech Republic. Therefore, complete restoration of oligotrophic streams is the key to the future presence and natural reproduction of freshwater pearl mussels in the Czech Republic.

1. Introduction

Freshwater species, particularly bivalve molluscs, are greatly threatened with extinction (Bauer and Wachtler, 2001; Strayer, 2006), and their protection requires a comprehensive approach (Geist, 2011). The freshwater pearl mussel (Margaritifera margaritifera) was

historically an abundant species in oligotrophic streams, but it has been disappearing rapidly in recent decades. This species has a complex life cycle that begins with a very brief planktonic phase in the form of microscopic glochidia, which is followed by a parasitic phase on the gills of salmonid fishes for nearly a year. The completion of metamorphosis is followed by a hidden phase in hyporheic zone. Finally, the young mussel emerges at the bottom surface and may live up to a century as a stable, sessile benthic filter-feeder consuming fine detritus (Bauer and Wächtler, 2001). The complexity of the freshwater pearl mussel's lifecycle makes it vulnerable to rapid anthropogenic changes that have affected freshwater ecosystems over the last two centuries.

Currently, the freshwater pearl mussel is designated endangered species based on European Union legislation and is protected within Natura 2000. According to the national law of the Czech Republic, the species has been listed as critically endangered since 1992 and has been under legal protection since 1913, when it was declared a year-round protected species and harvesting of wild populations was banned. Mussels were allowed to be cultured for pearl production in artificial millraces. The last documented official pearl hunt occurred in 1944 in the millrace of Otava River in Horazdovice (Dyk, 1947; Dyk and Dykova, 1974). The conservation of the freshwater pearl mussel in the Czech Republic was covered by the nation's general environmental protection legislation announced in 1955.

Despite their protection over several decades, the freshwater pearl mussel is retreating from its natural range across Europe (Araujo and Ramos, 2001; Popov and Ostrovsky, 2013). Generally, the causes of declines have been eutrophication, river regulations and changes in land use in catchments (Bauer, 1988; Hruska, 1999; Gumpinger et al., 2002; Geist, 2010). In the Czech Republic, the first documented massive die-out of freshwater pearl mussels occurred via the construction of paper mills and other industrial enterprises between 1850 and 1940 (Dyk, 1947). Subsequently, the discharge of untreated sewage water, the construction of dams, and the regulation of stream canals had additional profound negative impacts on this species. Land use changes also occurred; for example, traditional farming on less productive land was gradually abandoned such that former fields and meadows returned to natural succession, and agriculture on more productive fields was intensified (Dyk, 1992; Hruska, 1991b). Forest diversity was modified when the proportion of spruce monocultures increased. The water course within the landscape was managed via the massive construction of drainage systems, which caused the physical destruction of many natural streams due to digging and elevated levels of erosion (NCA CR, 2013).

These cumulative effects had a significant impact on the temperature regime, erosion patterns and water chemistry, which resulted in changes in the aquatic food webs of individual river basins. A distinct lack of detritus particles with sufficient nutrient and calcium content (Hruska, 1991a; Ticha et al., 2012) are considered to be the main factors preventing the reproduction and survival of juveniles in most of the residual Czech populations (Hruska, 1991a, 1991b; NCA CR, 2013). The absence of regular population recruitment led to population declines. Conservation efforts for remaining freshwater pearl mussel populations and research on the causes of reproductive failure began in the 1980s (Bauer et al., 1980; Young, 1991). The freshwater pearl mussel is currently recognised as a flagship species for the complex protection of oligotrophic catchments due to its importance as a sensitive species (Buddensiek, 1995; Hruska, 1999; Geist, 2010). Despite the conservation and protection efforts of this mussel species, few action plans in Europe have resulted in significant population increases or improvements in the age structure, apart from the Lutter River in Germany (Buddensiek and Ratzbor, 1995; Altmüller and Dettmer, 2006).

In the Czech Republic, systematic activities to protect the freshwater pearl mussels have been conducted since 1982 (Hruska, 1985), and a comprehensive action plan is currently being conducted. The main conservation and protection efforts focus on habitat protection combined with direct population support measures, including breeding mussels and providing optimal

conditions and special management in specially built side-arms of the river (Hruska, 1999).

The aim of this study is to summarise the current status of populations of freshwater pearl mussels in the Czech Republic compared with historical populations. We provide an overview of the population reinforcement over the past 25 years and the presence of juveniles in these populations. In addition, we discuss the effectiveness of the conservation measures and evaluate the initial results of the long-term attempts to augment the residual populations.

2. Methods

2.1 The study area

The study area is located on the southeast edge of the freshwater pearl mussel range in Europe. The area includes the following river catchments: the Vltava and Saale Basins of the upper Elbe River, which drains into the North Sea; the Oder and Nisa Basins of the upper Oder River, which drains into the Baltic Sea; and the Morava Basin of the Danube River, which drains into the Black Sea. The original range of the freshwater pearl mussel is relatively sparsely settled and is characterised by a high percentage of forested areas, with a current predominance of spruce monocultures. In the bedrock, crystalline rocks (granite, granodiorite and diorite) are dominant.

The conservation of freshwater pearl mussels in the study area is managed by the governmental Nature Conservation Agency (NCA CR) in the form of a national action plan. The action plan categorises localities with active species populations according to their current and predicted statuses (Supplement A). An ecosystem-oriented conservation approach is used to reflect the complex processes in oligotrophic waters that are key to freshwater pearl mussels' survival and reproduction.

2.2 Analysis of available data on historic freshwater pearl mussel occurrence

The data of the historical occurrence of freshwater pearl mussels in Czech rivers and streams were summarised from various reports (Schubert, 1933; Nowak, 1936; Dyk, 1947; Podubsky and Stedronsky, 1955; Svatos, 1971; Dyk and Dykova, 1974; Bauer, 1992; Flasar, 1992). Historical occurrences obtained from secondary sources were marked as unreliable records. Unfortunately, historical sources only considered the occurrence of freshwater pearl mussels in rivers and streams, rather than the quantitative estimates of the populations. In addition, because surveys before 1950 were conducted for economic purposes, minor populations were likely overlooked. Current populations located in small streams may also have been overlooked. Nevertheless, all current mussel localities were presumed to have had populations in the past.

2.3 Analysis of the present occurrence and abundance of freshwater pearl mussels

The freshwater pearl mussel is listed as critically endangered in the Czech Republic; thus, it is protected by a law that prohibits disturbing M. margaritifera in its natural habitat, handling mussels during any phase of their life cycle, manipulating individuals, extracting individuals from stable positions, and killing mussels. Therefore, research must follow the restriction under the conservation law. For example, screening of the habitat for the presence of mussels can be performed only by counting mussels visible on the bottom surface because disturbing their habitat by ransacking or sieving of the sediment (Hastie et al., 2004; Young et al., 2001) is illegal. These research restrictions limit the accuracy of population size and age structure estimates.

Current abundance data were obtained from the survey database of the action plan and the authors' observations. A summary of the knowledge regarding mussel occurrence was plotted in standard KFME squares (Ehrendorfer and Hamann, 1965).

Streams were usually inspected at the beginning of the growing season after the water

had receded from the elevated spring levels but before the foliage grew on bank-shore trees and before the periphyton overgrowth reduced the visibility within the stream. The abundance of pearl mussels in smaller, shallower streams, order II-IV according to Strahler (Strahler, 1957), with relatively lower numbers of individuals (Hastie and Cosgrove, 2002) was assessed by complete screening as counting from the shore or carefully wading directly through a particular stream. Using this approach, all mussels visible on the sediment surface were directly counted using an aquascope (0 12 cm) with additional lighting for shadowed microhabitats or using a magnifying glass for detecting juveniles. This methodology was repeatedly conducted in Blanice, Luzni potok and Bystfina (Fig. 1 localities No. 1 and 3). In the other locations, the complete screening approach was not used due to the habitat size. The occurrence of mussels was screened only in habitats that were presumed suitable, and shells and mussel remains were noted where mussels may have been overlooked upstream. Therefore, mussel abundances in larger habitats should be considered minimum estimates of the likely abundances.

We distinguished between the following two types of localities: localities with living individuals and extinct localities with no signs of freshwater pearl mussels. As such, we assessed localities with previously known freshwater pearl mussel populations but repeatedly found no individuals or only empty shells (Young et al., 2001). Reliable evidence of extirpation is available only at a few sites because the streams with severely altered habitats that are not expected to significantly improve are not priorities of the national action plan.

A reconstruction of the historical areal distribution for comparison with the current distribution of pearl mussel populations was conducted in South Bohemia (southwest of the Czech Republic), where the historical distribution was sufficiently described due to previous pearl fishing. We compared the occurrence in streams of order IV and higher according to Strahler because the data on smaller streams may have been underestimated.

Comparative analysis of the distributions of both historical and current localities based on stream-order flow (Strahler, 1957) and altitude was conducted using a GIS tool (MapInfo) at a map scale of 1:10,000. The term "locality" refers to a site with freshwater pearl mussel presence within a stream or stream section (the stream section should be at least 10 km in length; we used this approach for localities with continuous occurrences of mussles). Localities with a total abundance of less than 10 individuals and artificial habitats (millraces) were not included in the analysis. Elevations above sea level (m a.s.l.) were determined in the middle section of the stream section, and the data were then categorised into 50 m intervals. Because the historical data actually underestimated the occurrence of freshwater pearl mussels in streams of order III and lower, we used absence-presence records.

2.4 Analysis of the freshwater pearl mussel population genetic structure

To describe the genetic structure of Czech freshwater pearl mussel populations, we analysed 134 individuals at 10 localities: two locations at the upper Blanice (20 and 15 ind.), two locations at the lower Blanice (9 and 8 ind.), the upper Zlaty potok (16 ind.), the upper Malse (15 ind.), Tepla Vltava (15 ind.), Bystfina (15 ind.), Luzni potok (15 ind.), and Jankovsky potok (6 ind.). DNA was extracted using the DNeasy Blood & Tissue Kit (Qiagen) from the haemolymph of living adult and subadult individuals, except for one sample that was obtained from a deceased adult. Haemolymph was sampled non-invasively and under a special permit following the protocol of Geist and Kuehn (Geist and Kuehn, 2005).

All individuals were genotyped for 12 microsatellite loci (MarMa 1632, MarMa 2671, MarMa 3050, MarMa 3621, MarMa 4322, MarMa 4726, MarMa 5167, MarMa 5280, MarMa 3116, MarMa 4277, MarMa 4315, and MarMa 4859; Geist et al., 2003) using the Multiplex PCR kit (Qiagen), followed by a fragment analysis on an automatic sequencing machine (ABI 3130 Genetic Analyser, Applied Biosystems). The genotypes were scored using GeneMapper Software 3.7 (Applied Biosystems). To assess the relationship among individuals, we performed a factorial correspondence analysis (FCA) using the program Genetix 4.05.2

(Belkhir et al., 2004).

2.5 Age structure assessment

The laws forbidding the handling of mussels in their natural habitat prevented the possibility of construction of cohort diagrams and the exact determination of the age structure of Czech freshwater pearl mussel populations. However, living individuals were measured during emergency transfers due to habitat loss, such as drought or construction in streams. Such transfers were conducted recently at two localities, Blanice and Zlaty potok, in 2000 and 2002, respectively, on sets of 1408 and 87 individuals.

An individual assessment of the age of living freshwater pearl mussels was possible due to the knowledge of size-to-age relationships previously determined using deceased individuals. Shells of various sizes were collected from particular localities and were analysed using the ligament-sectioning technique according to Hendelberg (Hendelberg, 1961). The annuli were counted, and a regression curve specific to each locality was computed (archived by NCA CR) using the measured length (i.e., length of shell, length of ligament, and length of corroded ligament). Each living individual was measured with an accuracy of 0.1 mm (length of shell, length of ligament, and length of corroded ligament). By comparing the data obtained with the regression curve for each particular locality, we determined the age of the individual to a decadal accuracy.

The approximate dates of the ceased natural reproduction were based on cohort diagrams. Alternatively, we used knowledge of the last known occurrences of small individuals under 6.5 cm that were presumably juveniles (Y oung et al., 2001).

To estimate the proportion of juveniles in the current populations, the age classes (adult/subadult) of smaller mussels were assessed by inspecting the distance between the outer edges of exhalant and inhalant apertures. Individuals with both siphons within a 3 cm distance were included in the subadult category (Matasova et al., 2013). Because no overlap occurs in the sizes of adult and subadult mussels, we could distinguish younger mussels with a high level of accuracy.

2.6 Population reinforcement

Population reinforcement ofM. margaritifera was conducted in all remaining localities within the last three decades. To avoid genetic transfers between areas of population reinforcement, the identity of the local population was strictly respected, and only local genotypes for each particular locality were used. Czech population reinforcement efforts were conducted via two approaches: the release of glochidia-infected fish or the release of captive-bred juveniles (the ages varied between 3-5 years on average). The year(s) of the population reinforcements are summarised in Tables 1 and 2. In Blanice, Luzni potok and Zlaty potok, juveniles were released into specially built side-arms with special meadow management to provide optimal habitat conditions (Hruska, 1999).

3. Results

3.1 The historic occurrence of freshwater pearl mussels

The range of the historical occurrence of freshwater pearl mussels in the Czech Republic is depicted in Fig. 1. M.argaritifera margaritifera historically lived in the drainage areas of the North Sea, and abundant populations lived in the mountainous streams south of the Elbe Basin (Vltava, Saale). The most abundant populations were found in the catchment area of the Vltava (Moldau) in South Bohemia (southwest of the Czech Republic), primarily in large rivers, such as Otava to Pisek, Blanice, and Malse (Maltsch) and Vltava to Ceske Budejovice (Fig. 2). The species was also occasionally found in upper streams of the Odra Basin bordering Poland (northeast of the Czech Republic; drainage of the Baltic Sea; Tab. 1). These populations

(Luzicka Nisa: No. 5 in Fig. 1; Kladska Nisa No. 6 in Fig. 1) were isolated, sparse, and located at lower altitudes. These populations are currently extinct. The last known evidence of mussels living within the Luzicka Nisa catchment was in Kocici potok, where two individuals were found in 1940. In 1977, only shell remains were found. The last known populations in the Kladska Nisa Basin (Cerny potok near village Vidnava) had disappeared by 1991.

Other traditionally listed but uncertain historical occurrences of freshwater pearl mussels were in the Danube Basin (No. 7 in Fig. 1; southeast of the Czech Republic; Black Sea) and Elbe localities, such as Orlice, Doubrava and Chrudimka (No. 4, Central Czech Republic; Dyk, 1947).

3.2 The present status of freshwater pearl mussel populations

The current total size ofM. margaritifera populations in the Czech Republic was estimated as 16,000 individuals (Tab. 1). The freshwater pearl mussel has disappeared from all localities at altitudes below 500 m a.s.l. and from large rivers and streams (order V and higher). Therefore, most of the current localities can be found near the upper limit of the mussel's historical range at altitudes of approximately 700 m a.s.l. and in streams of order III-IV (Fig. 3).

The most abundant populations in South Bohemia, which is known in the catchment area of the Vltava River, disappeared from 90% of the localities between 1850 and 2012; anestimated 99% of the population died out (Fig. 2). The current populations are mainly preserved in river systems of Blanice, Zlaty potok and Tepla Vltava. Freshwater pearl mussels can also be found in the Malse River on the Czech-Austrian border. The second largest zone of freshwater pearl mussels in the Czech Republic is located in the western Saale Basin in the connected system of Luzni potok, Bystfina and Rokytnice on the Czech-Germany border.

A specific feature of the Czech mussel populations is the large number of localities at state borders in transboundary river basins (Fig. 1, Tab. 1), such as at Malse/Maltsch or other sites on the border of Germany. The Svarcava/Schwarzbach, Kamenny potok/Bieberbach and Kouba/Chamb streams flow from the Czech Republic to Bavaria, Germany where there are significant mussel populations (Fig. 1). Similarly, Rokytnice/Regnitz, Pekelsky potok/Hollbach and Ujezdsky potok/Mahringsbach host populations near state borders and downstream in Bavaria (Fig. 1).

In the Sazava Basin, central Czech Republic (i.e., Jankovsky potok and its tributary Kladinsky potok), mussels are threatened because of significantly altered habitats (see characteristics in Supplement A). The last fragment of the population has only 2 individuals, which were found during a population screening in 2012 (Tab. 1).

3.3 Genetic structure of populations

The genetic analysis of the current population structure showed three distinct groups, referred to as conservation units (CU), in the Czech Republic (Fig. 5). CU 1 contains the populations that inhabit the majority of the Vltava Basin (Blanice, Zlaty potok and Tepla Vltava) and, surprisingly, the population from the remote locality of the Sazava Basin (Jankovsky potok). CU 2 includes individuals from Malse that belong also to the Vltava Basin and some of the individuals originating from Tepla Vltava, where both CU 1 and CU 2 individuals are present (Fig. 2). CU 3 comprises mussels from the geographically distant localities of the Saale Basin in the western Czech Republic. This genetic structure was respected during population reinforcement efforts.

3.4 Age structure of Czech populations and efforts for population rejuvenation

In most of the Czech populations ofM. margaritifera, only adult and elderly individuals are present. Detailed information regarding age structure is available only for Blanice and Zlaty potok, which both show a lack of natural recruitment during the last three decades (Fig. 4). The most recent records of juveniles in the Czech Republic were as follows: upper Blanice, decline

from 1965 to 1977; lower Blanice, last recorded in 1952; upper Zlaty potok, between 1965 and 1975; and Kremezsky potok, 1971. In western localities, the last juveniles were recorded in Luzni potok between 1965 and 1975 and in Bystrina between 1970 and 1980. In summary, interruption of the reproductive cycle began in the third quarter of the 20th century.

Nevertheless, in recent years, we came across evidence of active (or recently active) natural reproduction in two localities. We found juveniles in the lower Blanice (in an artificial locality of a millrace in 2011; 15 ind.) and in the upper Malse (2012; 38 ind.) These localities were never supported by artificial rejuvenation efforts (see below).

The impaired age structure of the Czech mussel populations led to measures aimed at improving the proportion of the juveniles in populations with respect to local genotypes (see above). Juveniles and subadults are currently present in several populations that were reinforced via the release of infested trout and/or the release of grown juveniles reared through semi-natural breeding methods in recent decades (Tab. 1). Trout with attached glochidia on their gills were stocked in two localities: the upper Blanice and Tepla Vltava. In the upper Blanice, approximately 6,000 fish were released between 1984 and 1995; in total, the fish carried approximately 10 million glochidia (Tab. 2). Between 1999 and 2002, 415 trout with an average of 2,000 glochidia were released in Tepla Vltava, carrying in a total of 0.83 million freshwater pearl mussel glochidia.

Five localities (upper Blanice, upper Zlaty potok, upper Malse, Tepla Vltava and Luzni potok) were significantly enhanced by the release of grown juveniles. A total of 53,302 individuals were released, with a minimum of several hundred mussels in each location (Tab. 1). Two other localities (Jankovsky potok and Bystrina) received fewer juveniles, with tens of individuals at most (see Tab. 1 for detailed information).

After completing the hidden hyporheal life phase, juveniles emerge at the sediment surface in respective time. However, no cohort was found after the release of the infested fish in Blanice. Similarly, we did not see any results in Tepla Vltava yet. In contrast, we detected juveniles at the grown juvenile release sites at all reinforced localities. The most significant proportion of juveniles was observed in the upper Blanice and upper Zlaty potok, where younglings composed 10% of the population (Tab. 1). Because of the small size of mussels and their partially hidden life cycle, the estimated abundances are likely underestimated.

4. Discussion

4.1 Historic range of freshwater pearl mussels and possible transfers

The reconstruction of the original range of the mussels in the Czech Republic was based on a number of well-documented reports on their occurrence or on pearl fishing. The species was economically important, particularly in South Bohemia (Dyk, 1992); thus, the occurrence data presented here is considered valid. Nevertheless, less numerous populations or upstream populations were likely overlooked. This is evidenced by the fact that most of the current localities lack historical data.

Some records of M. margaritifera in the central Elbe Basin and in the Moravian area of the Danube Basin are uncertain (No. 4 and 7, Fig. 1). However, the streams flowing from the Danube-Elbe drainage divide to the Danube are inhabited by freshwater pearl mussels at a number of localities in northern Bavaria, Germany, and Austria (Geist and Kuehn, 2005; Gumpinger et al., 2002). In addition, reports of freshwater pearl mussel occurrence in the Morava Basin may have resulted from confusion with the thick shelled river mussel (Unio crassus), which is similar in morphology and still occurs in many river basins (Douda et al., 2012). Additionally, the Czech scientific name did not distinguish between the genera Unio and Margaritifera until the 1950s. Further detailed research on historical data is therefore needed.

A special feature of the occurrence of freshwater pearl mussels in Central Europe is that they are present despite the major drainage divide of the Elbe-Danube-Oder (Geist and Kuehn, 2005). One possible explanation is that the flow path within the watershed changed in the geological past (Geist and Kuehn, 2005). Nevertheless, some populations may have been established as a result of the deliberate transfer of infected fish or M. margaritifera individuals for subsequent pearl harvesting. In the Czech Republic, such profit-motivated attempts were documented within the basin of the Vltava River in the 18th and 19th centuries (Stepan, 1927 in Dyk and Dykova, 1974). Similar effects in the Middle Ages when fish-pond culturing and intense pearl fishing were frequent cannot be ruled out (Dyk, 1947). Therefore, some isolated Czech populations ofM. margaritifera could have come from human activities. This might explain why mussels from the geographically remote locality of Jankovsky potok (Fig. 1) are attributed genetically to the South Bohemian CU 1 Blanice. Further research on the genetic structure could determine whether the Jankovsky potok population was established through a few founders and could provide evidence of the founder effect documented in Waldaist, Austria, and other localities in Germany and Luxemburg (Geist and Kuehn, 2005). Alternatively, the population may have descended from a more abundant and widespread metapopulation in the past. A similar history was suggested for a number of currently small populations in the Danube Basin whose genetic markers indicated their natural origin (Geist and Kuehn, 2005). However, it may never be possible to reveal the true origin of freshwater pearl mussels in this locality due to their rapid population decline (Tab. 1).

4.2. Current range and fragmentation

Current populations of freshwater pearl mussels in the Czech Republic are located at high altitudes near the upper limit of the historical range and in minor upper streams (Figs 2, 3). Certain populations were likely maintained due to pronounced changes in the settlement after 1945 that significantly reduced human pressure on the ecosystem: a military training area is upstream of Blanice; Malse, Luzni potok and Bystfina are in the former Iron Curtain border zone. Elimination of human settlements allowed mussel populations to survive until conservation efforts were initiated in the 1980s (Hruska, 1985), while other once abundant populations in lowlands, such as lower Vltava and Otava (Fig. 2), completely died out in the first half of the 20th century.

The current range is also significantly fragmented. Rivers were transformed by damming (Fig. 2) and weirs that both block natural fish migrations (Musil et al., 2012). For example, salmon was a possible host for glochidia and was previously abundant in all localities with freshwater pearl mussels, except at Blanice (Fric, 1894). However, the first dam built on the lower Elbe in 1939 eliminated salmon occurrence in the Czech Republic. In addition to physical barriers in the streams, fragmentation of the habitat may arise from industrial and agricultural pollution (Dyk, 1992; Hruska, 1999). Poor conditions reduced the survival of trout and mussels and led to the separation of upstream and downstream localities.

Vltava populations (Blanice, Tepla Vltava, and Otava) were likely connected to other populations within the drainage divide of the North Sea (i.e., metapopulation occurrence). This hypothesis was supported by a microsatellite analysis when the relatively distant populations were clustered into one CU (Fig. 5). However, this system is no longer connected (Fig. 2). Western localities, including Luzni potok and Bystfina in the Saale Basin, which form one CU, are still interconnected. The Malse CU, which is relatively close to the Blanice CU but is distinct, is analogical to other localities found in Europe (Geist and Kuehn, 2005) where genetically unique populations are still present despite the connectivity of river systems and short distances.

4.3. Lack of natural reproduction

Long-term failure in completing the reproductive cycle is probably connected to poor habitat quality due to anthropogenic changes or the range limits of the species. For example, high altitudes (Figs 2, 3) with lower temperatures may prevent completion of the parasitic glochidial phase of the life cycle (Hruska, 1992).Two documented populations with scarce natural production of juveniles (lower Blanice and Malse) thrive in altered habitats,. Despite a small proportion of younglings present, the populations never increased in abundance or experienced improved age structures (Absolon and Hruska, 1999).

Nearly all current localities where M. margaritifera occur are in protected areas (Tab.1) that are intended to restrict the intensity of building, farming, timber industry and sewage water management (NCA CR, 2013). However, significant improvements in river health and declines in pollution in the last decade have not yet resulted in the restoration of the mussels' natural reproduction in the Czech Republic. At all localities, some parameters of the habitat are unsuitable for M. margaritifera, as defined by the action plan (Supplement A); therefore, passive protection of areas without active restoration of complex habitat features does not result in the desired improvement of the habitat quality (Maiorano et al., 2008; Geist, 2011; Laurance et al., 2012) that would subsequently lead to successful reproduction and juvenile survival.

4.4 Effectiveness of breeding efforts

The release of infected trout and grown juveniles is among the most frequently used techniques to enhance natural populations of freshwater pearl mussels (Gum et al., 2011). In this context, it is important to be aware of the quality of the environment at the release sites for survival of the youngest individuals. Experience and bioindicator tests from the Czech Republic showed that environmental quality meets the requirements of first stage juveniles at one location only (NCA CR, 2013). Thus there is no evidence of juvenile emergence from sediment at Blanice after 10 million glochidia (on trout) were released between 1984 and 1995, even though their proper development on gills was monitored by fish recapture (Hruska, 2000). Subsequent bioindicator tests at this site identified insufficient trophic conditions for the earliest and most sensitive life stages after metamorphosis and the beginning of hyporheal life (Hruska, 1999). Since then, infected trout have not been released at Blanice because the method is assumed to be inefficient. In contrast, at Tepla Vltava, where infested fish were released between 1999 and 2003, there are suitable conditions for the survival and growth of the youngest individuals, as suggested by bioindicator tests. Nevertheless, direct evidence of recruitment, such as the occurrence of juveniles and subadults (Matasova et al., 2013), is still missing. To evaluate the true effect of this measure, we must wait for the completion of the hyporheal life phase, which can take up to 20 years.

The release of juveniles (ages 3-5) showed more promising results. The population at Blanice was greatly enhanced as a result of the release of nearly 50,000 juveniles over 11 years into specially built side-arms in managed meadows adjacent to streams, which provided optimal habitat conditions for juveniles (Hruska, 1999). Similarly, in Zlaty potok, the proportion of subadults in the population has increased 10%, but the population rejuvenation is not yet sufficient to ensure the future persistence of freshwater pearl mussels. Therefore, the sole release of tens of thousands of grown juveniles that undergo natural mortality may not be adequate measure to keep freshwater pearl mussel in a particular streams and rivers. This highlights the urgent need for active habitat improvement that will allow natural reproduction in the near future; this improvement is included in the Czech action plan for freshwater pearl mussels.

5. Conclusions

All Czech populations ofM. margaritifera are highly skewed towards old individuals and fragmented altered habitats. Conservation efforts for preserving this species in the Czech Republic began approximately 30 years ago. The critical factors were identified as the poor

habitat quality, which is manifested in the failure of the complete reproductive cycle, and juvenile survival. Long-term breeding activities involving either the release of infected trout or grown juveniles have resulted in partial rejuvenation of several populations. Nevertheless, the proportion of young mussels is still below the levels of intact, self-recruiting populations. Based on these results, the Czech action plan has not yet been successful, and further significant improvements in environmental conditions are the key to the future survival and natural recruitment of the freshwater pearl mussel in the Czech Republic.

Acknowledgements :

O.P.S and M.B. were supported by grant from the Czech Ministry of Environment (MZP 0002071101). I.V. received institutional support RVO:60077344. Support for K.D. was obtained from the Czech Science Foundation (13-05872S) and ESF/MSMT (CZ. 1.07/2.3.00/30.0040). H.P. was supported by a grant from the EEA and Norway (grant no. 009/2). Data on the present occurrence and proportion of juveniles were collected during the run of the action plan for the freshwater pearl mussel (available at www.zachranneprogramy .cz) managed by the Nature Conservation Agency of the Czech Republic, which is funded by the government of the Czech Republic. We would like to thank the two reviewers whose valuable comments improved an earlier version of this manuscript. We would also like to thank Karel Absolon, Mojmir Elias, Ondrej Volf, Tereza Minarikova, Jan Svanyga, Ondrej Spisar, Denisa Blazkova, Richard Faina, Detmar Jäger, Eva Zelenkova and Bohumil Dort for their long-term cooperation during the action plan. Ondrej Spisar also assisted with the haemolymph sampling in the microsatellite analysis. Jan Vrba helped with the GIS analysis. We also thank Josef Rebec and Vojtech Mrazek for their support in the field work. Vera Kladivova helped gather the historical information.

References

Absolon, K., Hruska, J., 1999. Perlorodka ricni (Margaritifera margaritifera Linnaeus, 1758) v Ceské republice (Freshwater Pearl Mussel (Margaritifera margaritifera) Linnaeus, 1758 in the Czech Republic). NCA CR, Prague. 1-27. Altmüller, R., Dettmer, R., 2006. Successful species protection measures for the freshwater pearl mussel (Margaritifera margaritifera) through the reduction of unnaturally high loading of silt and sand in running waters. Inform. D. Naturschutz Niedersachs 26, 192-204.

Araujo, R., Ramos, M.A., 2001. Action plan for Margaritifera margaritifera in Europe.

Nature Environ. 117, 29-66. Bauer, G., 1988. Threats to the freshwater pearl mussel Margaritifera margaritifera L. in

Central Europe. Biological Conservation 45, 239. Bauer, G., 1992. Variation in the life span and size of the freshwater pearl mussel. Journal of

Animal Ecology 61, 425-436. Bauer, G., Schrimpf, E., Thomas, W., Herrmann, R., 1980. Relations between the decline of the freshwater pearl mussel (Margaritifera margaritifera) in the Fichtelgebirge and the water quality. Archiv für Hydrobiologie 88, 505-513.

Bauer, G., Wächtler, K., 2001. Ecology and Evolution of the Freshwater Mussels Unionida.

Ecological Studies 145, 1-394. Belkhir, K., Borsa, P., Chikhi, L., Raufaste, N., Bonhomme, F. GENETIX 4.05, logiciel sous Windows TM pour la génétique des populations, 2004. Montpellier (France).

Buddensiek, V., Ratzbor, G., 1995. Restoration of sedimental quality in a small brook of the Lüneburger Heide, Northern Germany. Folia 91, 19-24. Buddensiek, V., 1995. The culture of juvenile freshwater pearl mussels Margaritifera margaritifera L. in cages: A contribution to conservation programmes and the knowledge of habitat requirements. Biological Conservation 74, 33-40.

Douda, K., Horky, P., Bily, M., 2012. Host limitation of the thick-shelled river mussel:

identifying the threats to declining affiliate species. Animal Conservation 15, 536-544.

Dyk, V., 1947. Ceske perly (zivot, ochrana a narodohospodarsky vyznam perlorodek) (Czech pearls - life, protection and national economic importance of pearl mussels). Josef R. Vilimek, Praha. (in Czech).

Dyk, V., 1992. Profilovi ohrozovatele lokalit perlorodky ricni (Main threats to freshwater pearl mussel localities). Erica, Plzen 1, 21-38. (in Czech).

Dyk, V., Dykova, S., 1974. The pearl oyster (Margaritifera margaritifera Linnaeus 1758) a neglected indicator of the pollution of mountain and submountain water flews of the crystallone region in Czechoslovakia. Acta Veteriana 43, 287-304.

Ehrendorfer, F., Hamann, U., 1965. Vorschläge zu einer floristischen Kartierung von Mitteleuropa. Ber. Deutsch. Bot. Ges. 78, 35-50.

Flasar, I., 1992. Frühere Verbreitung der Flußperlmuschel (Margaritifera margaritifera (L.)) im Friedländer Gebiet in Nordböhmen (Eulamellibranchiata: Margaritiferidae). Malakologische Abhandlungen 16, 83-87.

Fric, A., 1894. Losos labsky - biologicka a anatomicka studie (Elbe salmon - A biological and anatomical study). Frantisek Rivnac, Praha. (in Czech).

Geist, J., 2010. Strategies for the conservation of endangered freshwater pearl mussels

(Margaritifera margaritifera L.): a Synthesis of Conservation Genetics and Ecology. Hydrobiologia 644, 69-88.

Geist, J., 2011. Integrative freshwater ecology and biodiversity conservation. Ecological Indicators 11, 1507-1516.

Geist, J., Kuehn, R., 2005. Genetic diversity and differentiation of central European

freshwater pearl mussel (Margaritifera margaritifera L.) populations: implications for conservation and management. Molecular Ecology 14, 425-439.

Geist, J., Rottmann, O., Schröder, W., Kühn, R., 2003. Development of microsatellite markers for the endangered freshwater pearl mussel Margaritifera margaritifera L. (Bivalvia: Unionoidea). Molecular Ecology Notes 3, 444-446.

Geist, J., Söderberg, H., Karlberg, A., Kuehn, R., 2010. Drainage-independent genetic structure and high genetic diversity of endangered freshwater pearl mussels in northern Europe. Conservation Genetics 11, 1339-1350.

Gum, B., Lange, M., Geist, J., 2011. A critical reflection on the success of rearing and culturing juvenile freshwater mussels with a focus on the endangered freshwater pearl mussel (Margaritifera margaritifera L.). Aquatic Conserv: Mar. Freshw. Ecosyst. 21, 743-751

Gumpinger, K., Heinisch, W., Moser, J., Ofenböck, T., Stundner, C., 2002. Die

Flusmussperlmuschel in Österreich. Umweltbundesamt - Federal Environmental Agency Austria, Wien.1-55.

Hastie, L.C., Cooksley, S.L., Scougall, F., Young, M.R., Boon, P.J., Gaywood, M.J., 2004. Applications of extensive survey techniques to describe freshwater pearl mussel distribution and macrohabitat in the River Spey, Scotland. River Research and Applications 20, 1001-1013.

Hastie, L.C., Cosgrove, P.J., 2002. Intensive searching for mussels in a fast-flowing river: an estimation of sampling bias. Journal of Conchology, 37, 309-316.

Hendelberg, J., 1961. The freshwater pearl mussel Margaritifera margaritifera (L.). Report of the Institute of Freshwater Research, Drottingholm, 41; 149-171.

Hruska, J., 1985. Ochrana perlorodky ricni v CHKO Sumava- jih. (Conservation of

freshwater pearl mussel in a southern region of Bohemian Forest protected landscape area). Pamatky a priroda 9, 559-562. (in Czech).

Hruska, J., 1991a. Projekt "Zachrana perlorodky ricni v Ceske republice" 2. cast Biotop perlorodky ricni (The freshwater pearl mussel conservation project in the Czech

Republic. Part 2: Freshwater pearl mussel biotopes). Pamatky a priroda 16, 609-612. (in Czech).

Hruska, J., 1991b. Projekt "Zachrana perlorodky ricni v Ceske republice" l.cast. Zakladni

Charakteristika a priciny ohrozeni (The freshwater pearl mussel conservation project in the Czech Republic. Part 1.: Basic characteristic and causes of a threat). Pamatky a priroda 12, 545-548. (in Czech).

Hruska, J., 1992. The freshwater pearl mussel in South Bohemia: Evaluation of the effect of temperature on reproduction, growth and age structure of the population. Arch. Hydrobiol. 126, 181-191.

Hruska, J., 1999. Nahrungsansprüche der Flußperlmuschel und deren halbnatürliche Aufzucht in der Tschechischen Republik. Heldia 4, 69-79.

Hruska, J., 2000 Experience of semi-natural breeding programme of freshwater pearl mussel in the Czech Republic. Ergebnisse des Kongresses vom 16.-18.10.2000 in Hof. Die Flussperlmuschel in Europa: Bestandssituation und Schutzmassnahmen , 69-75. 2000.

Laurance, W.F., Useche, C.D., Rendeiro, J., et al., 2012. Averting biodiversity collapse in tropical forest protected areas. Nature 489, 290-294.

Maiorano, L., Falcucci, A., Boitani, L., 2008. Size-dependent resistance of protected areas to land-use change. Proceedings of the Royal Society B: Biological Sciences 275, 12971304.

Matasova, K., Simon, O.P., Dort, B., Douda, K., Bily, M., 2013. Recent distribution of

freshwater pearl mussel (Margaritifera margaritifera) at historical localities in the upper part of the Vltava River basin (Czech Republic). Silva Gabreta 19, 139-148.

Moritz, C., 2002. Strategies to protect biological diversity and the evolutionary processes that sustain it. Systematic Biology 51, 238-254.

Musil, J., Horky, P., Slavik, O., Zboril, A., Horka, P., 2012. The response of young of the year fish to river obstacles: Functional and numerical linkages between dams, weirs, fish habitat guilds and biotic integrity across large spatial scale. Ecological Indicators 23, 634-640.

NCA CR, 2013. Zachranny program pro perlorodku ricni v CR (Action plan for the endangered freshwater pearl mussel in the Czech Republic). NCA CR, 1-76. (in Czech).

Nowak, W., 1936. Perlorodka ricni a jeji perly (se zvlastnim zretelem k pomerüm v republice Ceskoslovenske) (Freshwater pearl mussel and its pearls (with respect to the Czechoslovac Republic conditions). Sbornik vyzkumnych ustavü zemedelskych CSR 146, 1-76. (in Czech).

Popov, I.Yu., Ostrovsky, A.N., 2013. Survival and extinction of the southern populations of freshwater pearl mussel Margaritifera margaritifera in Russia (Leningradskaya and Novgorodskaya oblast). Hydrobiologia 735, 161-177.

Podubsky, V., Stedronsky, E., 1955. Doplnky k biologii perlorodky (Margaritifera margaritifera L.). (Supplements to Freshwater pearl mussel (Margaritifera margaritifera L.) biology) Ziva 3, 155-156. (in Czech).

Schubert, O., 1933. Über die Perlmuschel - und Perlen - Vorkommen in Böhmen. Landeskulturrat für Böhmen, Prag.

Strahler, A.N., 1957. Quantitative analysis of watershed geomorphology. Transactions, American Geophysical Union 38, 913-920.

Strayer, D.L., 2006. Challenges for freshwater invertebrate conservation. Journal of the North American Benthological Society 25, 271-287.

Svatos, I., 1971. Vyskyt perlorodky ricni na upati Blanskeho lesa (Freshwater pearl mussel

occurrence in the Blansky les mountains foothills). Ochrana Prirody 26, 220-221. (in Czech).

Ticha, K., Simon, O.P., Douda, K., Kubikova, L., 2012. Detrital Components in Submontane

611 Organogenic Springs in Relation to Their Morphology, Microhabitats and

612 Macroinvertebrates. Polish Journal of Ecology 60, 163-175.

613 Young, M.R., 1991. Conserving the freshwater Pearl Mussel (Margaritifera margaritifera L.)

614 in the British Isles and Continental Europe. Aquatic Conservation: Marine and

615 Freshwater Ecosystems 1, 73-77.

616 Young, M.R., Cosgrove, P.J., Hastie, L.C., Henninger, B., 2001. A standardised method for

617 assessing the status of freshwater mussels in clear, shallow rivers. Journal of Molluscan

618 Studies 67, 395-396.

619 Zelinka, M., Marvan, P., 1961. Zur Präzisierung der biologischen Klassifikation der

620 Reinheit fließender Gewässer. Archiv für Hydrobiologie 57:389-407.

FIGURES Captions

Figure 1: Comparison of the historical and present occurrence of freshwater pearl mussels in the Czech Republic and the recent population reinforcement. The network of standard KFME squares and main river systems is indicated. Recently extinct and uncertain historic localities of M. margaritifera are as follows: No. 4 Orlice, Doubrava, Chrudimka; No. 5 Luzicka Nisa/Lausitzer Neisse, Ploucnice; No. 6 Kladska Nisa/Nysa Klodzka; and No. 7 Becva. Current populations are labelled according to their affiliation with the conservation units (CU). No. 1 Blanice CU comprises populations in Blanice, Zlaty potok, part of Tepla Vltava (southwestern Czech Republic) and Jankovsky potok (central Czech Republic). No. 2 comprises Malse/Maltsch and part of Tepla Vltava. No. 3. Saale comprises populations in Luzni potok/Zinnbach and Bystfina/Wolfsbach. More details regarding the CUs are given in Fig. 5. Population reinforcement with grown juveniles according to their numbers is indicated by the symbol size (triangles). The arrows indicate the streams of Svarcava/Schwarzbach, Kamenny potok/Bieberbach and Kouba/Chamb (Danube Basin) that flow from the Czech Republic to Bavaria and Germany, where significant populations still occur.

Figure 2: Detailed overview of the range reduction of the M. margaritifera in the southwest part of the Vltava River Basin between 1850 and 2013. The presence of mussels is indicated with a black line or points in streams that are order IV and higher (according to Strahler, 1957). The triangles show constructed dam reservoirs.

Figure 3: Altitudinal distribution of freshwater pearl mussel localities in the Czech Republic with respect to stream order, according to Strahler. Historic extinct localities (open circles) and current localities (full circles) are indicated. The term locality refers to a site where M. margaritifera is present in a stream or a 10 km section of a stream where localities are continuous at longer distances. The elevation was determined in the middle of the transect and then categorised into 50 m intervals.

Figure 4: Age cohort diagram (using the ligament-sectioning technique) for two Czech freshwater pearl mussel populations. The data were obtained during emergency transfers of the populations in Zlaty potok in 2002 (Nind=87) and in Blanice in 2000 (Nind=1408). The data originated from the archive of NCA CR (J. Hruska).

Figure 5: The genetic structure of the Czech freshwater pearl mussel population as determined by microsatellite analysis. The results of the factorial correspondence analysis are shown, and each genotyped individual is displayed as a point in a two-dimensional space, which is defined by two factorial axes that represent factors explaining the majority of the variability in the given dataset. The individuals analysed form three clusters, referred to as conservation units (CU). CUs are labelled as follows: No. 1 refers to Blanice CU, No. 2 is Malse CU, and No. 3 is Saale CU. The numbers correspond to Figure 1.

TABLES

Table 1: Overview of the recent localities, population sizes and reinforcements of freshwater pearl mussels in the Czech Republic. Border streams and recently extinct localities were included. For levels of legal protection, the following abbreviations were used: EVL (EC Habitats Directive Natura 2000 protected area); I-IV (national protected area according to the IUCN classification), BG (Biogenetic Reservation of the Council of Europe id CZ930001). For population reinforcement, F is fish infestations (for upper Blanice see details in Table 2) and the numerical value is the number of released grown juvenile mussels. For population screening, the year and discovered size of the mussel population (Nind.) are indicated; * indicates that the method of complete screening could not be used due to the habitat size (see Methods). The number of juveniles (out of all individuals) was underestimated because we only counted individuals visible on the sediment surface; in some localities, this information was not available (n.a.).

Table 2: Summary of the release of fish (Salmo trutta m. fario) infected with glochidia of M. margaritifera at the upper Blanice between 1984 and 1995.

Table 1

Legal protection Population Population Population

Basin Local basin River Locality Country reinforcement (Nind, year) screening (year) size (Nind) Subadults

Blanice upper CZ EVL, IV, BG F(1984 -1995); 49,468 2010 10,120 > 1 %; locally

(1995-2005) > 10 %

Blanice Blanice lower & Zlaty potok lower CZ EVL, IV 0 2011 358 > 1 %

Zlaty potok upper CZ EVL 887 (20022003) 2005 1710 > 10 % 184 ind.

Dluhost'sky potok CZ IV 0 2011 0 0

Vltava (CZ) Chvalsinsky potok CZ IV 0 2006 6 n.a.

Kremzsky potok CZ IV 0 2005 50 n.a

Elbe (CZ/DE) Vltava Malse upper Malse lower CZ/AU CZ EVL, IV EVL 438 (2005) 0 2012 1996 > 440* 202 > 10 % 49 ind. n.a.

Tepla Vltava CZ EVL,II F (1999-2003); 1,180 (1998) 2013 > 331* > 1 % 18 ind.

Sazava Jankovsky potok & Kladinsky potok CZ EVL,IV 42 (2005) 2012 2 2 ind.

Bystrina CZ/DE EVL, III 34 (2003) 2009 594 n.a.

Rokytnice upper CZ/DE EVL, IV 0 2013 1 1 ind.

Saale (DE) Luzni potok CZ/DE EVL, IV 1,329 (19952001) 2013 2,034 1 % 19 ind.

Ùjezdsky potok CZ/DE - 0 2012 12 0

Pekelsky potok CZ/DE - 0 2012 0 0

Odra (CZ/PL) Odra (PL) Klodzka (PL) Lausitzer Cerny potok Kocici potok CZ CZ/PL 0 0 1992 1977 0 0 0 0

Neisse Smêda (CZ/PL/DE) Smeda CZ - 0 1977 0 0

Table 2

Year 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 Total

Number of host fish 665 600 235 400 609 320 690 443 430 510 484 423 5,809

Average N glochidia

1.5 1.5 3.2 3.0 1.3 3.0 3.5 1.0 1.5 2.0 2.0 2.0 2.125

per fish (ind. 103)

Proximate total N

glochidia (ind. 106) 1.00 0.90 0.75 1.20 0.79 0.96 2.42 0.44 0.65 1.02 0.97 0.85 10.09

^T o oo ^

M recent occurrence in 1990-2013 # historic occurrence before 1990 O uncertain historical occurrence KFME quadrates watercourse T 100 - 1.000 juveniles ▼ more than 10.000 juveniles

I NXTlI "T'Tlii

o o o o o

/;.fA000 «» / \

O O 7 O O OO

N ^_____

«wr* 0 ¿U/K 2013. O HCA CR 2014. C VUV1GM 2014

100 km

Before 1850

Figure 3

ACCEPTED MANUSCRIPT

900 800 r? 700 CD 600 — 500

<D ■O

3 400 < 300 200 100

Stream order

o 0 o •

• o • • o

• • o

o o o •

o • • o o

• o o o

o o o o o 0

o o o o o

o o o

o 0 o o

o o o o

T->-I-1-'-I-r

re 3 30

<u Q. 10

Zlaty potok

45 40 35 30 25 20 15 10 5 0

Blanice

10 20 30 40 50 60 70 80 90 Age cohort (years)

10 20 30 40 50 60 70 80 90 100 110 >110 Age cohort (years)

-1.5 L--1.5

OBIanice upper □ Blanice lower AZIaty potok upper + Jankovsky potok ▲ Tepla Vltava O MalSe upper X Bystrina — Luzni potok

0.0 0.5 1.0 1.5