Abstract My main objectives in this thesis were to explore general patterns and mechanisms driving salmonid invasions globally and, more specifically, to examine the invasion dynamics and impacts of the North American brook trout in North European stream systems. Non-native salmonids have often spread extensively and caused many harmful impacts on their native counterparts. Among the three globally introduced salmonids, the European brown trout appeared as the 'worst' alien species (strongest impact on native fish), followed by the North American rainbow trout and brook trout. Brook trout, which is widely established in European streams, was found to be a non-aggressive species. Moreover, the growth of brown trout was unaffected by brook trout, indicating negligible interspecific interactions between the two species. Habitat niche segregation between brook trout and brown trout was evident, with brook trout occupying mainly low-velocity pool habitats, whereas brown trout resided in fast-flowing riffles. At the river-wide scale, brook trout occurred mainly in small, slightly acid headwater streams, whereas brown trout was found primarily in larger downstream river sections. Evidently, North European streams with a very low number of native fish species offer underutilized niche space for tolerant headwater specialists such as brook trout. However, the habitat niche filled by brook trout was not completely vacant, as brown trout co-occurred with brook trout in numerous small and mid-sized (3–16 m wide) streams. In these streams, brown trout reproduction was negligible presumably related to the presence of brook trout. Brook trout had also increased in density relative to brown trout during the 10-yr study period (1994 vs. 2004). Moreover, the growth rate and population densities of brook trout were high and the species had spread extensively towards the upmost headwater streams during the 10-yr study period. Thus, harmful effects on the native brown trout by brook trout are likely to occur in the core habitat of the invader, i.e. headwater streams, leaving populations of the native species unaffected elsewhere. Due to the high conservation value of the potentially impacted populations of brown trout, I strongly caution against further stocking of brook trout in European watersheds.

Current literature suggests that little, if any, research has been conducted in South Africa to determine the impact of alien invasive trout on indigenous amphibian biodiversity. The primary aim of the research project was to establish whether waterfalls are seasonally important in protecting the indigenous Natal Cascade Frog Hadromophryne natalensis tadpole populations from predation by alien rainbow trout Oncorhynchus mykiss and brown trout Salmo trutta at two sampling areas located at Injesuthi and Monk‟s Cowl Nature Reserve within the uKhahlamba Drakensberg Park World Heritage Site (UDPWHS). The dissertation argues in favour of a biotic rationale, namely trout predation, as being the primary cause for the decline of H. natalensis tadpoles below the waterfalls and systematically negates the influence of geo-physical (abiotic) environmental factors on tadpole abundance. Habitat isolation and fragmentation is identified as a latent threat to the continued persistence of H. natalensis populations in the UDPWHS. / Environmental Sciences / M. Sc. (Environmental Science)

This study aimed to evaluate otter predation on stocked trout and grayling. The diet composition of piscivorous Eurasian otters (Lutra lutra) was studied on the stream Chotýšanka in central Bohemia using spraint (otter faeces) analysis during one winter period (90 days). Stocking took place mostly in the second half of 2005. Around 5000 hatchery-reared brown trouts (Salmo trutta m. fario) of size about 10 cm, 600 graylings (Thymallus thymallus) of the same size and 480 rainbow trouts (Oncorhynchus mykiss) of size about 30 cm were stocked into the stream. Spraints were collected in December 2005 and January 2006. We found and identified 1531 fish in the spraints. The primary food source of otters was fish (almost 85 % of biomass). The main fish prey was small - sized, below 20 g in weight and 13 cm in length. The bulk of otter's diet consisted of less - valued species, especially the gudgeon (Gobio gobio). Other frequently consumed fish species were the European chub (Squalius cephalus), the common carp (Cyprinus carpio), the European perch (Perca fluviatilis), the brown trout and the rainbow trout. Consumption of stocked brown trout was 1,80 %, while cosumption of stocked rainbow trout was 10,90 % of biomass. Otters in this area consumed about 723 - 1446 brown trouts and 72 - 144 rainbow trouts...

In the present study an amplification and genotyping system for 9 microsatellite loci has been developed and used to genotype brown trout (Salmo trutta) individuals. This system has allowed, on one hand, to efficiently analyze the brown trout population structure from Catalan Pyrenees basins and, on the other, to describe their genetic demography. The complexity of evolutionary processes occurred in the area and the different scale of space-time in which they took place make difficult to draw a general pattern relating structure and genetic demography of brown trout populations in the Catalan Pyrenees. However, results fit with the population structure suggested for the Mediterranean brown trout populations, which consist of small interconnected demes forming metapopulations. In the studied Pyrenean basins, the reproductive strategy and the connectivity between demes mitigate the genetic drift expected from the estimated effective population sizes. / El sistema de 9 loci microsatélites desarrollado en este trabajo resulta muy eficiente para el análisis de la estructura poblacional y la demografía genética de las poblaciones de trucha común (Salmo trutta) en las cuencas del Pirineo. La complejidad de los procesos evolutivos ocurridos durante las glaciaciones del cuaternario en el Pirineo y la diferente escala espacial y temporal en que estos han tenido lugar impiden establecer un patrón general que relacione la estructura poblacional y la demografía genética en las poblaciones de trucha común del Pirineo catalán. Sin embargo, los resultados obtenidos están de acuerdo con la estructura poblacional sugerida para las poblaciones mediterráneas de trucha común, que se consideran constituidas por pequeñas agrupaciones locales interconectadas que constituyen metapoblaciones. En todas las cuencas del Pirineo parece que hay una tendencia a que la estrategia reproductora y la interconexión entre los grupos limiten los efectos perjudiciales de la deriva.

There are five endemic populations of migratory Atlantic salmon (Salmo salar, Salmonidae) and brown trout (Salmo trutta, Salmonidae) remaining in Lake Vänern, whereof two are salmon and three are trout. There is one population of each species in Klarälven, also one of each in Gullspångsälven and there is a population of trout still remaining in Tidan River. Since the beginning of the 20th century endemic populations of salmon and trout have either vanished or declined dramatically due to the construction of hydropower in the rivers around the lake, and still today the future of the remaining populations remains uncertain. To manage salmonid fishes effectively more knowledge of their ecology throughout their different life-stages is needed. A large part of the Lake Vänern salmonoids lives are spent in the lake, where a mayor mortality factor is recreational fishing, especially angling. Estimation of fishing effort and catch is important in order to remain a sustainable fishery. In Lake Vänern commercial fishers are bound to report their catch, but anglers are not, -even though they are now thought to be responsible for the largest proportion of the catch. Angler interviews and access point surveys are the best ways to get accurate information in order to assess sport fish effort and catch. There are drawbacks of this method though that includes large costs, logistical problems and biased results due to the higher chance of interviewing the most avid anglers. To hand out mail-surveys is a kind of off-site survey that is less costly compare to interviews, but the drawbacks are bigger in getting correct information due to biased results from non-respondence. A study was conducted to compare an interview-survey with a mail-in survey, in which the interviews were conducted in Lillängshamnen, a port on the northern coast of Lake Vänern, and the mail-in surveys were distributed in different ports around the lake, though not in Lillängshamnen. This study shows a significant difference in catch per boat day between the interviews and mail-in survey. To find out more about how to develop the interview-survey there was also an analysis done on the correlation between weather and the number of interviews that were collected. One could find a negative correlation between numbers of interviews collected and the wind ratio, where very few interviews were collected once the wind speed measured more than 5-8 meters per second. There are improvements to be made both for the conduction of interviews and mail-in surveys, including more staff for conducting interviews at different ports and follow-ups on the non-respondents for the mail-in survey. / Av atlantlax (Salmo salar, Salmonidae) och öring (Salmo trutta, Salmonidae) finns idag fem endemiska populationer kvar i Vänern varav två är av lax och tre av öring. Det finns en klarälvsvariant av båda arterna, likaså finns det en gullspångsvariant av varje och sen finns även en öringpopulation kvar i Tidan. Sen början av 1900-talet har populationer av lax och öring utrotats eller minskat kraftigt till följd av vattenkraftens utbyggnad i vattendragen runt Vänern, och än idag är de resterande populationernas framtid oviss. För att kunna förvalta laxfiskarna på ett bra sätt krävs kunskaper om deras ekologi i dess olika livsstadier. En stor del av laxfiskarna runt Vänerns liv tillbringas i Vänern, där en mortalitetsfaktor består av fritidsfisket, då främst trolling. Uppskattning av ansträngning och fångst är viktigt för att hålla fisket i sjön på en hållbar nivå. Yrkesfisket måste rapportera all sin fångst men detta krävs inte av trollingfisket som idag tros stå för den största proportionen lax- och öringsfångst i Vänern. Genomförandet av intervjuer är en typ av plats- och åtkomstpunktsundersökning och är den bästa metoden att samla in korrekt information för att uppskatta sportfiskets ansträngning och fångst. Det finns dock nackdelar i form av stora kostnader, svår logistik och vinklade resultat då de mest entusiastiska fiskarna löper större chans att bli intervjuade. Att dela ut enkäter är en form av distansundersökning och även brevundersökning som kostar betydligt mindre än att genomföra intervjuer men brister vad gäller svarsfrekvens som i sin tur leder till vinklade resultat. En studie gjordes för att jämföra intervjuer med enkäter, där intervjuer genomfördes i Lillängshamnen på Vänerns norra kust och enkäter delades ut i olika hamnar runt hela Vänern, dock ej i Lillängshamnen. Denna undersökning visade att enkäter har en signifikant större fångst per båt än intervjuer. För att ta reda på mer om hur genomförandet av intervjuer skulle kunna förbättras undersöktes hur vädret påverkade antalet intervjuer som genomfördes. Man fick en negativ korrelation mellan antalet genomförda intervjuer och vindhastighet där ytterst få intervjuer genomfördes i vindhastigheter över 5-8 meter per sekund. Det finns förbättringspotential vad gäller både genomförandet av intervjuer och enkätundersökningar, bland annat mer personal till att kunna intervjua fler folk i olika hamnar och uppföljning på de enkäter som inte svarats på.

Se ha analizado las causas de la distribución espacial de la variabilidad genética del ADN mitocondrial en poblaciones de trucha común de la cuenca del Duero y de los Pirineos Orientales. En total se han analizado de novo 49 localidades, 13 en la cuenca del río Duero y 36 en los principales ríos del Pirineo oriental. Además se analizaron las fluctuaciones temporales en 14 de las localidades del Pirineo Oriental. Estudios previos indican un marcado contraste de los patrones de diversidad entre ambos territorios.En la cuenca del río Duero los análisis confirmaron la presencia de los dos linajes matriarcales descritos previamente, el linaje Atlántico (AT) y el linaje Duero (DU). Los análisis de la varianza molecular (AMOVA) siguiendo una jerarquía hidrográfica sugirieron una alta estructuración de las poblaciones coincidente con los patrones ictiológicos observados en la cuenca. El linaje DU parece haber estado presente permanentemente en la cuenca interior del Duero, mientras que las zonas más próximas a la desembocadura han padecido diversas colonizaciones de trucha del linaje AT, que reflejarían los cambios climáticos ocurridos en el Cuaternario. Se ha detectado una discrepancia en el límite entre ambos grupos definidos por genes nucleares (alozimas) y el ADN mitocondrial. Estas discrepancias pueden ser debidas a un efecto más severo de la deriva genética en el ADN mitocondrial que en los marcadores nucleares. Sin embargo, en este trabajo se han observado evidencias a favor de selección en el ADN mitocondrial del linaje DU que también explicaría estas discrepancias.El análisis más exhaustivo en las cuencas de los Pirineos orientales, permitió detectar nuevos haplotipos mitocondriales de los linajes Adriático (AD) y Mediterráneo (ME). En esta región, los AMOVAs confirmaron que las diferencias entre poblaciones dentro de río son más importantes que las diferencias entre ríos. No obstante se observó un patrón de aislamiento por distancia en toda la zona, reflejo de la estructuración de las poblaciones en la cuenca del río Ebro. Además, aunque los AMOVAs mostraron que el componente temporal de la variación es inferior al espacial, las fluctuaciones temporales en la comparación matriarcal de las poblaciones resultaron estadísticamente significativas. Estas fluctuaciones están asociadas tanto a la deriva genética como a procesos de flujo génico entre poblaciones próximas. Dentro de las cuencas, los componentes de diferenciación entre afluentes son, en general, superiores a los obtenidos dentro de cada afluente, patrón que parece estar extendido en la trucha común. Los estudios a escala microgeográfica en la Noguera Vallferrera y Noguera Cardós (afluentes del Noguera Pallaresa) reprodujeron este patrón de diferenciación. Los tamaños efectivos y la tasa de migración entre ambos ríos fueron similares a los descritos en poblaciones noratlánticas. Los tamaños efectivos de las hembras (Nef), calculados a partir del ADN mitocondrial fueron menos de la mitad del tamaño efectivo total tanto en la Noguera Vallferrera como en el resto de localidades pirenaicas estudiadas. Estos bajos tamaños efectivos de las hembras serían también responsables de las fluctuaciones temporales observadas. Los ejemplares repoblados parecen hibridar poco con los nativos, pero su presencia podría intensificar indirectamente los procesos de deriva genética y complicar la conservación de los patrimonios genéticos nativos.Con la salvedad de la existencia de selección que favorece a los haplotipos del linaje DU, los procesos poblacionales que regulan la distribución de la variabilidad genética en la cuenca del Duero y en los Pirineos Orientales podrían ser parecidos y caracterizados por la existencia de múltiples demes interconectados a lo largo del curso fluvial. / Brown trout populations from the River Duero basin and from Eastern Pyrenees rivers were analyzed to assess the reasons for contrasting patterns of genetic diversity. Altogether genetic diversity has been analyzed in 49 new collections, 13 from the River Duero basin and 36 from the main rivers of Eastern Pyrenees. Moreover, temporal samples from 14 Pyrenean locations were sampled to analyse temporal stability of the described structure. In these two areas previous studies indicated a strong contrast among diversity patterns in both territories. Results in the Duero basin confirmed the presence of the Atlantic (AT) and the Duero (DU) matriarchal lineages both previously described in this river basin. The analyses of molecular variance (AMOVA) based on the hydrographical hierarchy indicated a high level of population structuring, in accordance with the icthiological pattern observed in this basin. The DU lineage permanently occupied the internal area of the River Duero basin, whereas zones close to the mouth of the river have suffered diverse waves of colonisations of trout belonging to the AT lineage, which would reflect the changes happened in the Quaternary. Discrepancies in the limits between both groups defined by nuclear genes (allozymes) and mitochondrial DNA have been detected. These discrepancies could be due to a more intense effect of genetic drift in mitochondrial DNA than in nuclear markers. Nevertheless, evidences in favour of selection in the mitochondrial DNA of the DU lineage have been described in this work, which also would explain this discrepancy. A detailed analysis of brown trout populations from rivers in the Eastern Pyrenees detected new mitochondrial haplotypes of the Adriatic (AD) and the Mediterranean (ME) lineages. In this region, the AMOVAs indicated that differences between populations within river were larger than differences between rivers. Nevertheless, a pattern of isolation by distance was observed in the whole zone, reflecting population structure within the River Ebro. The AMOVAs showed that the temporal component of the variation is lower than the spatial component, but the temporal fluctuations in the matriarchal comparison of the populations were statistically significant. These fluctuations were associated to both genetic drift and gene flow among close populations. Generally in the river basins, higher differentiation between than within stream was observed. This pattern seems to be widespread in brown trout. The studies on microgeographical scale undertaken in the Noguera Vallferrera and Noguera Cardós (tributaries of Noguera Pallaresa) reproduced the above pattern of differentiation. Effective population sizes and migration rate between both rivers were similar to those described in North-Atlantic populations. In the Noguera Vallferrera as well as in the rest of Pyrenean populations, the female effective sizes (Nef), calculated from mitochondrial DNA were less than a half of the total effective sizes detected. These low female effective sizes also contribute to the observed temporal fluctuations. Hatchery individuals hybridise poorly with the native one, but its presence could indirectly intensify genetic drift and complicate the conservation of the native genetic resources.In spite of selection favouring haplotypes of the DU lineage, population processes controlling the distribution of genetic variability in the Duero and the Eastern Pyrenees river basins could be similar and characterized by the existence of interconnected multiple demes throughout the fluvial course.

Durant anys, el principal mètode de gestió de les poblacions de truita comuna (Salmo trutta L.) ha estat la repoblació amb exemplars exògens. El seguiment genètic de les poblacions de truita comuna dels Pirineus orientals, realitzat en aquest tesi, indica que els al.lels procedents d'aquestes repoblacions estan conduint a una homogeneització de les poblacions naturals i a la pèrdua de la seva història evolutiva. D'aquí la importància de la detecció de la introgressió en el desenvolupament de noves estratègies de gestió i conservació de les poblacions d'aquesta espècie. En aquest treball, s'ha avaluat l'eficàcia de diferents marcadors i mètodes que ens ofereix la genètica de poblacions en la detecció de la introgressió present a les poblacions naturals. Alhora que s'ha analizat la influència que han tingut les reserves genètiques, aplicades amb posterioritat a les repoblacions, i que intenten equilibrar l'explotació i la conservació dels recursos genètics de les poblacions natives. / For many years, the primary strategy to maintain or increase brown trout (Salmo trutta L.) population size has been stocking with exogenous individuals. Results on genetic monitoring of brown trout populations in the eastern Pyrenees, carried out in this thesis, indicates that hatchery alleles are diluting native gene pools, leading to a genetic homogenization of native populations and the loss of evolutionary history of this specie. In this context, it is very important the continuous monitoting of the levels of introgression in wild populations to properly develop management and conservation strategies for populations of this specie. In this work, the efficiency of different genetic markers and methods for detecting introgression in natural populations has been evaluated. Moreover, we evaluated the influence of the new management approaches, genetic refuges, aimed to avoid introgression and preserve native genetic diversity remaining in wild populations.

Current literature suggests that little, if any, research has been conducted in South Africa to determine the impact of alien invasive trout on indigenous amphibian biodiversity. The primary aim of the research project was to establish whether waterfalls are seasonally important in protecting the indigenous Natal Cascade Frog Hadromophryne natalensis tadpole populations from predation by alien rainbow trout Oncorhynchus mykiss and brown trout Salmo trutta at two sampling areas located at Injesuthi and Monk‟s Cowl Nature Reserve within the uKhahlamba Drakensberg Park World Heritage Site (UDPWHS). The dissertation argues in favour of a biotic rationale, namely trout predation, as being the primary cause for the decline of H. natalensis tadpoles below the waterfalls and systematically negates the influence of geo-physical (abiotic) environmental factors on tadpole abundance. Habitat isolation and fragmentation is identified as a latent threat to the continued persistence of H. natalensis populations in the UDPWHS. / Environmental Sciences / M. Sc. (Environmental Science)

Bien que l’écologie de la truite (Salmo trutta) ait déjà été bien étudiée, les processus expliquant la dynamique spatio-temporelle des populations restent à caractériser. L’objectif de cette thèse était d’identifier les processus démographiques (ex. survie ou déplacements) et l’influence des processus biotiques (interactions entre individus) et abiotiques (conditions environnementales) qui structurent dans le temps et dans l’espace les différentes classes d’âge (alevins, juvéniles et adultes) de ces populations. J’ai étudié l’influence sur la survie apparente de (1) la densité-dépendance et (2) des conditions environnementales vécues directement par les truites (habitat hydraulique et température de l’eau). Pour évaluer la transférabilité des résultats, j’ai évalué (3) l’échelle spatiale (globale ou locale) à laquelle opéraient les processus et (4) si ces processus variaient entre populations (approche hiérarchique). J’ai considéré la dynamique de 45 populations de truites dont 22 sont situées à l’aval d’un ouvrage hydroélectrique. A large échelle, il est apparu que le recrutement de populations séparées par des distances allant jusqu’à 75km peut être synchronisé par de fortes crues lors de l’émergence des alevins ou des déplacements du substrat de ponte. Nous avons synthétisé les résultats de l’application d’un modèle déterministe de dynamique de population, calibrés localement sur neuf stations aux conditions environnementales bien caractérisées. Cette synthèse a montré que des processus locaux influençaient directement la dynamique des populations. Enfin, la construction d’un modèle hiérarchique a montré le rôle structurant de la mortalité densité-dépendante des juvéniles et des adultes, dont l’intensité augmentait en l’absence d’abris (<2% de la surface) ou variait avec la température de l’eau (diminution pour les juvéniles et augmentation pour les adultes). Ce travail fournit des bases scientifiques aux gestionnaires d’ouvrages hydro-électriques pour leur permettre de limiter leur influence sur les populations de poissons et répondre ainsi aux demandes réglementaires / Brown trout (Salmo trutta) ecology was largely studied. However, the process structuring the spatiotemporal patterns of population dynamics remains unclear. The objective of this thesis was to identify the demographic processes (e.g. survival or displacements) and the influence of biotic (between-individuals interactions) and abiotic (environmental conditions) processes structuring the age-stages (fry, juveniles and adults) of these populations in time and space.I studied (1) the role of density-dependence on survival and (2) the influence of environmental conditions experienced by trouts (hydraulics and water temperature). To assess the results’ transferability, I studied (3) the spatial scale (global or local) of influence of the processes and (4) if those processes varied among populations (hierarchical approach).I considered the dynamics of 45 trout populations, 22 being located downstream a hydropower facility. At a large scale, the recruitment of distant populations (up to 75 km apart) may be synchronized by large floods during emergence of fry or by spawning substratum displacements. We summarized the results of a determinist population dynamics models, locally calibrated on nine reaches showing well-described environmental conditions. This summary revealed the influence of various local drivers on population dynamics. Finally, a hierarchical model showed that density-dependent mortality among juveniles and adults was a key biotic process. The strength of the competition was greater in absence of shelter (<2% of the surface) and varied with water temperature (decreasing for juveniles and increasing for adults).The results of this work will provide scientific basis to hydropower facility managers. This will help them to reduce their influence on trout populations and respond to regulatory demands

Swimming performance strongly influences fitness in aquatic organisms and is closely tied to external body morphology. Although this connection has been closely examined at the individual and species level, few studies have focused on this relationship as it pertains to functional group assemblages. Using functional groups based on similarities in habitat use and morphology, I tested the hypothesis that swimming performance can be reliably predicted by functional group composition. I measured swimming performance as burst speed using a simulated predator attack and as prolonged speed using a step-endurance test in a laboratory flume. I measured morphology using geometric morphometric techniques. A difference in swimming behavior in four of the seven species was observed in the step-endurance test. Benthic species exhibited bracing behavior as an alternative to body-caudal fin (BCF) propulsion in the prolonged speed trials. Swimming performance exhibited a weak relationship with functional groups based on habitat or morphology. Rather a species-based model was the best predictor of swimming performance. Although species exhibited variation in swimming performance, body size was the strongest predictor of absolute swimming performance across all models. Relative swimming performance (measured in body lengths/sec) was negatively related to body size. The results of this study suggest that functional groups are not always reliable predictors of performance and they necessitate empirical testing to validate their effectiveness. This study also provides critical swimming performance data for previously unstudied Great Basin fishes which could be valuable for predicting fish passage through culverts, weirs and fish ladders.