Climate change impacts on production and dynamics of fish populations

Per Hedström

Departement of Ecology and Environmental Science Umeå 2016

This work is protected by the Swedish Copyright Legislation (Act 1960:729) ISBN: 978-91-7601-604-6 Front cover: Sven Norman Elektronisk version tillgänglig på http://umu.diva-portal.org/ Tryck/Printed by: KBC Service Center, Umeå University Umeå, Sweden 2016

You will never complete, and that’s as it should be… .. Whithin each of them vault after vault opened endlessly.

Tomas Tranströmer, from Romanesque Arches

iv

i

Table of Contents Table of Contents iAbstract iiList of Chapters iiiAuthor contributions ivBackground 1

Effects of temperature on ecosystem productivity and fish populations 1Effects of increased run off of terrestrial organic matter on ecosystem productivity and fish populations. 3Effects of temperature and increased run off of terrestrial organic matter on ecosystem productivity and fish populations. 4Aim of this thesis 5

Study system and methods 7Ecology of the study species 7Ethical approval 7Experimental system 7Abiotic variables 9Estimates of primary production 9Resources 10Fish 10

Results and discussion 11I: Recruitment in a changing environment 11II: Winter survival 12III: Population responses to climate change 13IV: Effects of warming on fish population demographics 15

Concluding remarks 17References 20Tack 28

ii

Abstract Ongoing climate change is predicted to increase water temperatures and export of terrestrial dissolved matter (TDOM) to aquatic ecosystems influencing ecosystem productivity, food web dynamics and production of top consumers. Ecosystem productivity is mainly determined by the rates of primary production (GPP) in turn controlled by nutrients, light availability and temperature, while temperature alone affect vital rates like consumption and metabolic rates and maintenance requirements of consumers. Increased level of TDOM causes brownification of water which may cause light limitation in algae and decrease GPP and especially so in the benthic habitat. Temperature increase has a been suggested to increase metabolic rates of consumers to larger extent than the corresponding effect on GPP, which suggest reduced top consumer biomass and production with warming.

The aim of this thesis was to experimentally study the effects of increased temperature and TDOM on habitat specific and whole ecosystem GPP and fish densities and production. In a replicated large-scale pond experiment encompassing natural food webs of lotic ecosystems I studied population level responses to warming and brownification in the three- spined stickleback (Gasterosteus aculeatus).

Results showed overall that warming had no effect on whole ecosystem GPP, likely due to nutrient limitation, while TDOM input decreased benthic GPP but stimulated pelagic GPP. In fish, results first of all suggested that recruitment in sticklebacks over summer was negatively affected by warming as maintenance requirements in relation to GPP increased and thereby increased starvation mortality of young-of-the-year (YOY) sticklebacks. Secondly, brownification increased mortality over winter in YOY as the negative effect on light conditions likely decreased search efficiency and caused lower consumption rates and starvation over winter in sticklebacks. Third, seasonal production of YOY, older, and total stickleback production was negatively affected by warming, while increased TDOM caused decreased YOY and total fish production. The combined effect of the two was intermediate but still negative. Temperature effects on fish production were likely a result of increased energy requirements of fish in relation to resource production and intake rates whereas the negative effect of TDOM likely was a result of decreased benthic resource production. Finally, effects of warming over a three-year period caused total fish density and biomass and abundance of both mature and old fish to decrease, while proportion of young fish increased. The main cause behind the strong negative effects of warming on fish population biomass and changes in population demographic parameters were likely the temperature driven increased energy requirements relative to resource production and cohort competition.

The results from this thesis suggest that predicted climate change impacts on lentic aquatic ecosystems will decrease future densities and biomass of fish and negatively affect fish production and especially so in systems dominated by benthic resource production.

iii

List of Chapters

This thesis is a summary and discussion of the following four papers, referred to in the text by their respective Roman numerals:

I. Hedström, P., Rodríguez, P., Karlsson, J., Vasconcelos, F. R. and Byström, P. Warming but not increased terrestrial doc has negative effects on fish recruitment. Submitted manuscript

II. Hedström, P., Bystedt, D., Karlsson, J., Bokma, F. and Byström, P. Brownification increases winter mortality in fish. Accepted manuscript, Oecologia

III. Hedström, P., Rodríguez, P., Karlsson, J., Vasconcelos, F. R. and Byström, P. Population and size dependent responses in fish production to climate change. Manuscript.

IV. Byström, P., Hedström, P., Hotchkiss, E., Rodriguez, P., Vasconcelos F.R. and Karlsson, J. Warming decrease fish population densities and biomass. Manuscript.

iv

Author contributions

Chapter I

PH, PB and JK designed the study, PH, PB, PR and FRV performed field sampling, PH compiled data and performed the statistical analysis and wrote the first draft of manuscript. All authors contributed to the revision of the manuscript.

Chapter II

PH and PB designed the study, DB did the winter sampling and laboratory work, PH, PB, did the summer sampling prior and after winter, PH wrote the first draft of manuscript and did the statistical analysis, FB performed the Bayesian modelling on size dependent mortality, all co-authors provided editorial advice for the manuscript.

Chapter III

PH, PB and JK designed the study, PH, PB, PR and FRV performed field sampling, PH compiled data and performed the statistical analysis and wrote the first draft of manuscript. All authors contributed to the revision of the manuscript.

Chapter IV

PB, PH and JK designed the study, PH, PB, PR, EH, FRV performed field sampling, PB and PH compiled data and performed the statistical analysis and PB wrote the first draft of manuscript. All authors contributed to the revision of the manuscript.

1

Background

Most aquatic ecosystems are facing changes in drivers of ecosystem properties as current climate change predictions for e.g. northern Hemisphere suggest both a temperature increase of 3 - 5 °C and increased terrestrial input of dissolved organic carbon (Woodward et al. 2010, IPCC 2013). Hence, the future conditions determining food web characteristic’s and the ecosystem services that aquatic ecosystems provide today may look very different from present conditions.

Traditionally, important ecosystem services like fish production have been assumed to be determined by the rates of primary production at the base of the food web (Sterner et al. 1997, Rand and Stewart 1998, Ask et al. 2009a, Seekell et al. 2015). While primary production is directly controlled by inorganic nutrients, light availability and temperature (Sterner et al. 1997, Müren et al. 2005, O’Connor 2009, Karlsson et al. 2009, Cross et al. 2015), temperature alone also strongly affects all levels in aquatic food webs due to its impact on consumer vital rates like consumption and metabolisms (O’Connor 2009, Yvon-Durocher et al. 2010, Sibly et al. 2012, Ohlberger 2013). That is what this thesis is about - how the impact of climate change affects primary production relative to consumer energy requirements and how the net effects of the both translates into changes in fish biomass and production.

Effects of temperature on ecosystem productivity and fish populations

In the context of historical changes in climate, the present rates and magnitude of temperature change is considered rapid and strong and believed to have major impact on the functions of the ecosystems in general and aquatic ecosystems in particular due to the strong dominance of ectothermic organisms (Woodward et al. 2010). Temperature sets strong constraints in the basic environmental conditions for organisms. Increasing temperature causes an overall increase in maintenance requirements and changes in temperature are therefore expected to directly affect organisms across all trophic levels with strong and complex effects on food webs. For fish, at the top of the food web, temperature determines consumption capacities and metabolic rates and thereby also the fish populations top-down control on resource levels (Hairston 1993, Hansson et al. 2013). As temperature also strongly determine energy requirements to sustain growth and reproduction at the individual level (Brett 1979, Wootton 1984, Jobling

2

1995) temperature will have strong effects on biomass and production at the population level in fish (Ohlberger et al. 2011).

Already in 1997, Beisner et al. experimentally showed that Daphnia-algae systems were destabilized at higher temperatures and that this effect was further magnified in the presence of a top predator causing decreased densities or extinctions of the predator. Increased temperature has also theoretically been shown to decrease biomasses of both resources and consumers and destabilize population dynamics, the former due to increased metabolic losses and the latter caused by the assumption that increased search rate increases faster than resource production (Vasseur and McCann 2005). Furthermore, since heterotrophic metabolism has been suggested to scale faster with temperature than autotrophic production (López-Urrutia et al. 2006, O’Connor et al. 2011 and references therein), increasing temperatures will theoretically cause herbivore production and biomass to decrease with increasing temperature (O’Connor et al. 2011). Fussmann et al. (2014) also suggested that although warming may stabilise predator prey dynamics it will on the same time decrease carrying capacity of prey, decrease predator densities or cause extinction of the latter. Furthermore, Ohlberger et al. (2011) showed in a recent model analysis that increasing temperature may cause increased resource competition, slower growth, smaller sized fish and destabilized population and community dynamics as a consequence of that metabolic costs increase faster with temperature than resource production. In all, most theoretical approaches thus point to that increasing temperature should lead to a lower biomass and production of higher trophic levels. Experimental studies focusing on temperature effects on primary production and responses at higher trophic levels in general support theoretical predictions. In pelagic systems warming generally decreases or has no effect on herbivore biomass (Petchey et al. 1999, Strecker et al. 2004, Müren et al. 2005, O’Connor 2009, but see Hansson et al. 2013). Similarly, in benthic systems increasing temperature has little effect on or decreases abundance or production of macroinvertebrates (Baulch et al. 2005, Feuchtmayr et al. 2007, Tixier et al. 2008, Eriksson Wiklund et al. 2009). In more general terms: warming leads to stronger increase in metabolism than resource dependent feeding rates which in turn causes growth rates in consumers to be more sensitive to changes in resource levels(Vucic-Pestic et al. 2011, Ohlberger 2013). Moreover, this negative effect of warming may be further magnified when passing optimal temperature for feeding rates in consumers (Englund et al. 2011). Experimental studies including fish as top consumer are rare but more importantly results are more inconclusive. Mesocosm studies including a non-dynamic fish consumer, i.e. a few fish added as top consumer testing for top down effects and growth responses in fish show both positive(Lefébure et

3

al. 2013) and negative effects on herbivores biomass (Kratina et al. 2012, Hansson et al. 2013) and that short term growth rates of fish do not change (Lefébure et al. 2013) or increase with warming (Hansson et al. 2013). Finally, when allowing for fish population dynamics although still in relative small mesocosms results suggest that warming may cause hump shaped density relationships (Šorf et al. 2015) or reduced densities of fish although this result was mainly explained by oxygen deficiencies (Moran et al. 2010). Thus, by including more realism and resemblance of natural ecosystems, predictions from controlled experimental studies of the effects of warming on fish populations becomes less congruent. Moreover, above studies have been done in relative small tanks with water volumes less than a few cubic meters rendering population size of fish to be very low. With that in mind, predictions from these studies should be treated with precaution as fish population performance are likely to be influenced by random factors, enclosures effects and lack of larger scale natural ecosystem feed backs.

Effects of increased run off of terrestrial organic matter on ecosystem productivity and fish populations.

The expected increase in terrestrial dissolved organic matter (TDOM) due to increased runoff from the terrestrial watershed is expected to affect the aquatic ecosystems in several ways. First, increased concentrations of terrestrial dissolved organic carbon (TDOC) causes brownifications of the water which reduces light penetration and may cause light limitation for primary producers and especially so for benthic algae (Diehl et al. 2005, Jansson et al. 2007, Ask et al. 2009b). Hence, brownification may cause both general decrease in primary production but also change the relative importance of pelagic versus benthic primary production as main energy source for consumers (Karlsson et al. 2009, Bartels et al. 2016). Secondly, the TDOM often contains additional nutrients (N and P) which may stimulate pelagic primary production (Jansson 1998, Klug 2005, Rodríguez et al. 2016). Finally, TDOM is an additional allochthonous carbon source for bacteria in addition to the autochthonous carbon produced by pelagic algae (Tranvik 1988, Jansson et al. 2007).

For visual predators like fish increased brownification may negatively affect search efficiency of prey (Jönsson et al. 2011, Estlander et al. 2012). Hence, consumption rates in fish may decrease with increasing TDOC concentrations, which in turn affect growth rates negatively. Moreover, as benthic primary production decrease with brownification, cause a shift in the size structure of prey resources for fish. A decreased benthic primary

4

production reduces the availability of larger sized macroinvertebrates that graze algae directly on the sediment and shift prey size structure available to fish towards a dominance of small sized zooplankton in the pelagic habitat (Hecky and Hesslein 1995, Karlsson and Byström 2005). This shift in size structure of available prey for fish will increase competition for zooplankton where smaller fish are known to be better competitors (Persson 1986, Persson et al. 1998). Thus, brownification likely benefits small sized and juvenile fish which in turn may lead to decreased densities and production of adults and large sized fish (Reichstein et al. 2015).

Hence, whether or not the effect of increased input of TDOM will cause biomass and production of fish to increase or if the net effect is negative depends on the relative strength of above listed effects. Comparative field studies along a gradient of DOC concentrations render support for a negative effect and suggests that decreased benthic production can explain the observed negative correlations between fish population biomass and production and DOC concentrations and light extinctions in lakes(Karlsson et al. 2009, 2015). Experimental support for negative effects on fish are scarce, if any, and the few studies present have studied effects of TDOM on pelagic food webs only (Hansson et al. 2013, Lefébure et al. 2013). While Lefebure et al. (2013) reported no effects on TDOM on fish production, Hansson et al. (2013) concluded that TDOM may increase fish production. However, their conclusions were based on growth responses of a few individual fish only. Thus, by not accounting for ecosystem feedbacks at the population level and ignoring potential negative effects on benthic production, the conclusion from these studies may not hold in more natural scenarios when including the benthic habitat.

Effects of temperature and increased run off of terrestrial organic matter on ecosystem productivity and fish populations.

The above discussed changes in either temperature or TDOM inputs alone have each strong implication for top consumer responses and dynamics and ecosystem properties. Based on the similar directions in expected population responses in fish to each factor alone a reasonable hypothesis is that the negative effects will amplified and stronger. Warming and increase in TDOM will not occur independently of each other, although warming is expected to increase first and increased TDOM follow due to increased precipitation and terrestrial primary production and the thawing of permafrost (IPCC 2013). Still, despite the fact that warming and increased TDOM will occur

5

simultaneously there are only a few studies that have addressed the synergistic effects of both on fish populations biomass and production and aquatic ecosystem function. Both Hansson et al. (2013) and Lefébure et al. (2013) showed experimentally in a pelagic system that the combined effect of the two may cause an increase in zooplankton production and increased individual growth of planktivorous fish. While Lefébure et al. (2013) suggested that the increase in zooplankton production was due to increased bacterial production, Hansson et al (2013) showed that increased primary production was the main reason. Thus, both these studies suggest positive effects (or less negative effects) of the combined effect of warming and TDOM on fish production than either factor alone. However, these results again are based on individual growth responses of a few fish, thus not accounting for ecosystem feedbacks at the population levels or the expected negative effects on benthic primary production on fish production. Thus, studies including the benthic habitat and quantifying the effects of both warming and TDOM at the population level in fish are lacking and are likely necessary to provide more realistic predictions of overall climate change effects on fish populations.

Aim of this thesis

The overall aim of this thesis it to study the impacts on fish population biomass and production of the major predicted effects of climate change i.e. warming and brownification. More specifically, I wanted to identify general mechanism and drivers behind specific population responses to be able to predict the future effects of climate change in aquatic ecosystem on fish production. Although, all studies were performed within one ongoing large scale experiment we were able to experimentally study different aspects of population level responses to climate change. The first two chapters focus on the two key periods and main factors affecting recruitment in fish, the growth and survival over the first summer respectively the survival over winter, while two last chapters focusing on climate change impact on fish populations.

More specifically in each chapter I wanted:

I: to study the effects of climate change on recruitment mimicking a system where adults migrate into a shallow spawning and nursery area for YOY fish to spawn. We quantified the effects of increased temperature and TDOM on gross primary production and measured diets, conditions and sizes of YOY

6

fish. Finally, we estimated net effects on the final densities of the YOY cohort born in spring.

II: to study the effects of climate changed induced brownification on the consumption of resources and the survival over winter in YOY fish. We quantified changes in light and oxygen levels and sampled resources for fish over winter. Diets and changes in condition of fish during winter was measured and related to changes in size distributions and the survival of YOY over winter.

III: to study the effects of warming and TDOM alone or in combination on the net production of biomass of fish over a growth season at the population level. We quantified the effects of increased temperature and TDOM on habitat specific pelagic and benthic gross primary production (GPP) and resource abundance for fish. Net biomass changes over summer in total biomass and in YOY and older fish separately were quantified. We then related the climate change induced response in habitat specific primary production and emerging effects on resource abundances for fish to the measured response in total fish production and in YOY fish and older.

IV: to study the long term effects of warming on densities, biomass and dynamics of fish populations. Over three seasons we quantified the effects of warming on whole ecosystem GPP and resource abundance for fish. Effects of warming on total population densities and biomass development and on population demographics (YOY fish, older fish and density of mature fish) were quantified and related to the effects of warming on GPP and resource abundances.

7

Study system and methods Ecology of the study species

The three-spined stickleback (Gasterosteus aceulatus) is a common species in both freshwater and marine systems and the family of Gasterosteidae is represented around the northern hemisphere between about 35° to 70° north (Gross 1978, Wootton 1984). Sticklebacks are omnivorous and feed on both zooplankton and benthic macroinvertebrates (Wootton 1984). Generation time is relative short 1- 4 years (Wootton 1984) and was mainly 1 year studies used in this thesis. The short generation time is beneficial for the identification of treatments effects on population level at the temporal scale of this project. Female stickleback can produce several clutches of eggs at optimal conditions and fecundity is positively size dependent (Wootton 1973). Considering temperature effects on stickleback physiological capacities the optimal temperature for growth has been estimated to be 21,7°C and the highest temperature for positive growth is around 29°C (Lefébure et al. 2011). For search efficiency or attack rate Lefébure et al. (2014) shows that attack rate is an increasing function of temperature up to 21- 24 °C.

Ethical approval

Sampling methods, collection of experimental fish, method of sacrifices and design of experiment in this thesis comply with the current laws of Sweden and were approved by the local ethics committee of the Swedish National Board for Laboratory Animals in Umeå (CFN, license no. A-20-14 to Pär Byström).

Experimental system

All studies were performed at the Umeå University Experimental Ecosystem Facility (EXEF). EXEF is a large scale experimental pond system (73 m long, 23m wide with a depth of 1.6 m) divided into 20 enclosures (11.5 × 6.7 m) situated at (63° 48′ 34″ N, 20° 14′ 33″ E) close to Umeå university. EXEF allows for semi-natural ecosystem studies that span annual and inter-annual times scales including yearly natural ice and snow cover during winter seasons. Each enclosure contains an ecosystem with a soft bottom benthic habitat and naturally occurring benthic and pelagic primary producers,

8

invertebrate consumers and a reproducing three-spined stickleback population as a fish top consumer. All enclosure has separate in- and outlets for water and the facility allows for manipulations of input water characteristics, including warming with heat exchangers for eight of the enclosures to a predetermined level above ambient temperature during the ice free season. The ongoing studies at EXEF were initiated in May 2012, with the main aim to study whole ecosystem responses to climate change. This involved two treatments: increased temperature (Warm) and humic river water input (Humic)/terrestrial dissolved organic matter (TDOM) as experimental manipulations in a factorial design. From May to October, eight enclosures are heated 3° C above the natural temperature development (ambient temperature) in the eight other enclosures. Water from each warm enclosure is circulated through an individual heat exchanger and back to same enclosure. Separate temperature sensors in one of the ambient (i.e. natural season-dependent temperature development) and one heated enclosure continuously controls the closed flow system of heated media from an air-source heat pump (to each of the individual heat exchangers). The temperature difference (ΔT) between the heating media in the closed flow system and the flow of water from each warm enclosure is approximately 3°C. To mimic climate-change increased input of TDOM to freshwater systems, four heated and four ambient enclosures received inputs of natural humic water continuously (4 L min-1 in 2012 and 2014 and 8 L min-1 in 2013). The humic water was transported from a mid-sized stream 23 km north-east of EXEF. The other eight enclosures received similar water volumes of clear water from a groundwater source.

Fig. 1 Air view of the EXEF facility with its 20 enclosures. Dark enclosures are TDOM treatment, warm enclosures are the eight enclosures to the right whereas the eight ambient temperature enclosures lie to the left. The four enclosures in the middle served as temperature buffer zone and other experimental purposes.

9

Abiotic variables

Water temperature (°C) and light intensity (PAR, Photosynthetic Available Radiation, [umol m-2, sec]) were continuously monitored at 0.8 m water depth with temperature (TH2-F, UMS Germany) and light sensors (SQ-110, Apogee USA) and recorded with loggers (Delta-T Devices, UK). Minute-values for temperature and light were averaged over every 15 minutes all year around. For light extinction, measured the vertical change in PAR weekly at 25 cm increments with a Li-250A radiometer equipped with a spherical quantum sensor Li-193SA (Li-Cor, Lincoln, USA). The extinction coefficient, Kd was calculated as the slope of the linear regression of the natural logarithm of PAR versus depth (Dring 1984). Water samples were taken weekly from each enclosure, filtered, and stored at -20°C for later analysis of concentrations of total phosphorus (TP) and total nitrogen (TN). Samples for DOC determination were stored at +5°C, filtered through pre-combusted (400°C, 3h), acid washed Whatman GF/F filters and acidified with HCl before the analysis. TP was determined according to Murphy and Riley (1962) after digestion of the sample with potassium persulfate. DOC and TN were analysed using an IL 550 TOC/TN analyser (Hach-Lange GmbH, Dusseldorf, Germany).

Estimates of primary production

Estimations of GPP in benthic and pelagic habitats, were performed in each enclosure using 48 h in situ incubations. For pelagic incubations a 2 L (32 cm height, 8.6 cm inner diameter) transparent acrylic tubes were positioned vertically 50 cm below the water surface. For benthic incubations a 12 L semi-spherical transparent polycarbonate chambers (34 cm inner diameter) were deployed at the bottom surface. The chambers were equipped with a metal frame underneath assuring tight grounding to the sediment surface. Oxygen (O2) loggers (MiniDOT, PME, Vista, CA, USA), recording O2 concentration every 1 min, were positioned inside both the tubes and chambers during incubations. For estimates of whole ecosystem GPP metabolism, the O2 loggers were instead deployed 50 cm below the water surface for one week periods at the end of June, middle of July and in September 2012 and from 2013 and onwards continuously over the whole growth season. Habitat specific and whole ecosystem GPP were calculated from the changes in oxygen concentration (ΔO2) over time (Δt). For details on calculations see (Hotchkiss and Hall 2014, Jonsson et al. 2015, Rodríguez et al. 2016).

10

Resources

Every three weeks we sampled for zooplankton with a zooplankton net (100 µm, 25 cm diameter) by a 133 cm vertical pull through the water column. Zooplankton samples were conserved in Lugol’s solution. Macro invertebrates were sampled every six weeks by a benthic sample with a 30 cm wide bottom net scraped 70 cm. Macroinvertebrates were stored in ethanol. Analysis of the sampled resources. In the laboratory, macroinvertebrates and zooplankton were classified to order, family, or genus. Biomass was estimated using individual body lengths transformed to dry biomass with length-weight regressions (Dumont et al. 1975, Botrell et al. 1976, Persson et al. 1996).

Fish

In May 2012 40 adult sticklebacks were introduced to each enclosure. The sticklebacks were caught in May 2012 in a shallow spawning bay (63°45'14" N, 20°32'18" E) 18 km east of EXEF. Based on a subsample of sticklebacks (n=100), the male: female sex ratio was 46:54 and based on egg counts in females the initial population reproductive output was estimated to be 2655 eggs in each enclosure. These sticklebacks reproduced and YOY cohorts born in spring 2012 in each enclosure constitutes the origin of the stickleback populations which were studied throughout this thesis. Twice a year, late April-early May and Early October, stickleback population densities, size structure and biomass were estimated by seine-netting each enclosure three to five sequential times with a specially designed net. All captured fish per seine haul were placed in white barrel and photographed from above and thereafter released back to the enclosure after the final seine-netting effort. Number of fish and individual length were estimated by photo image analysis technique and population densities were estimated used the repeated removal method (Zippin 1956). Subsamples of fish from spring and autumn sampling were kept to analyse length dependent body weights mass, ingested biomass, diet composition, isotopic signals, otoliths for age determination. Additional samplings of fish throughout the growth seasons were also done with hand nets for additional data on diets and conditions of sticklebacks. For the study of winter survival in paper II, sampling for abiotic data, resources and fish followed the same protocol as summer sampling with the difference that sampling during experimental period was carried out through a hole in the ice early February, March and April.

11

Results and discussion I: Recruitment in a changing environment

Key results from the initial introduction of 40 adult sticklebacks and the fate of their reproduction were that increased temperature had as strong negative effect on fish recruitment while TDOM addition which doubled DOC concentration compared to controls did not affect biomass and number of recruits. Both these results stand in contrast to expected negative effects of increased DOC levels (Karlsson et al. 2009, 2015) and the suggested positive effects of increased temperature in temperate systems on recruitment (Hendersson and Brown 1985, Sponaugle et al. 2006, Ohlberger et al. 2011).

The estimated reproductive output of introduced females was around 2650 eggs, which was slightly lower than the numbers of YOY densities present in the treatments with ambient natural temperature development at the end of this experiment. This suggests that observed differences in recruitment levels at end of the experiment was not constrained by the initial reproductive output. The lack of negative effects of increasing DOC levels on recruitment in this study can likely be related to that the reduction in light was not strong enough to decrease overall primary production to levels where resource production for YOY sticklebacks became more constrained than what was the case with warming.

Our experimental system is also rather shallow and with the DOC levels around 8 mg L-1 and Kd values around 2 m-1, 6 % of incoming light still reached the bottom. Corresponding Kd value in clear water was around 0.8 m-1 and 30% of incoming light reach the bottom. However here it should be noted that a further increase in TDOM and thus a higher increase in DOC will cause stronger light extinctions likely to affect benthic production negatively likely to affect recruitment negatively. This was also observed the following year in the experimental system where a further increase in DOC concentrations to approximately 12 mg L-1 and Kd values around 2.5 m-1 caused negative effects on recruitment (chapter III). Light extinction to benthic habitat similar to the increased DOC concentration and hence Kd values will obviously also occur with increasing depths (Seekell et al. 2015). It is often the very shallow areas of lakes and marine systems that are the habitats for spawning and nursing of juveniles (Byström et al. 2003, Sundblad et al. 2014). Thus, the expected negative effects of increasing DOC concentrations on light conditions and resource production for juvenile fish may not occur in such shallow habitats, at least under only moderate increases in DOC concentrations, where depth is insufficient for the negative

12

effects of brownification to emerge.

In contrast, increased temperature had strong negative effects on recruitment levels, both in terms of numbers and total biomass of YOY sticklebacks. At the individual level, high growth rates in fish requires high temperatures but high individual growth rate will only occur if resources are abundant (Ohlberger, 2013). Hence, the relationship between high individual growth rate and recruitment depends on a combination of resource levels and temperature. As warming did not cause any increase in YOY fish growth and the fact that body condition and stomach fullness were lower in warm enclosures suggested that resource limitation increased for YOY stickleback in the warm treatments. With no change in primary production as observed in this study, metabolic theory predicts decreased carrying capacity and decreased abundance of consumers with temperature (Vasseur and McCann 2005, Allen et al. 2005, Ohlberger et al. 2011, O’Connor et al. 2011, Fussmann et al. 2014). Thus, the expected increase in individual growth with increased temperature could not be realized because resource production did not keep pace with the increased metabolic demand and consumption capacity of sticklebacks. Instead, warming likely caused increased starvation mortality and thus the lower recruitment levels observed in this study. Small scale experimental studies also show corresponding negative effects of warming on consumer biomass (Petchey et al. 1999, Strecker et al. 2004, Baulch et al. 2005, O’Connor 2009) as do longer term mesocosm experiments including a few number of fish have shown that warming negative effects on fish densities (Moran et al. 2010, Kratina et al. 2012, Šorf et al. 2015).

In conclusion, as the very shallow and near shore habitats in both lakes and marine ecosystems supports resources for temporarily high densities of growing juvenile fish, result from this study is highly relevant but also provide rather pessimistic predictions for the future function and productivity as nursery areas of these habitats. Hence, our results suggest that ongoing climate change warming will cause shallow nursery areas for fish to be less productive with negative effects on fish population densities.

II: Winter survival

While chapter I suggested no effect of increasing DOC levels on recruitment over summer, this study showed that increasing DOC levels instead increased the mortality of YOY sticklebacks over winter. In an already strongly light limited environment due to ice and snow cover, increased DOC

13

concentrations coincided with significant lower ingested prey biomass and body condition in YOY sticklebacks. As resource availability was similar (zooplankton) or even higher (chironomids) than under ambient conditions this strongly suggested that DOC increased brownification of water caused the observed decrease in both food intake and condition in sticklebacks. Strong and negative size dependent winter mortality in YOY fish is common in many species (Toneys and Coble 1979, Post and Evans 1989, Byström et al. 1998, Biro et al. 2004) and both decreased resource abundance and poor light conditions have been suggested to be main mechanisms for observed starvation mortality (Byström et al. 2006). Poor light conditions affect foraging rates negatively by decreasing reactive distance and capture success (Vogel and Beauchamp 1999, Helland et al. 2011, Ulvan et al. 2012, Jönsson et al. 2013). Similarly, brownification should cause similar effects but experimental evidence for negative effects on fish foraging rates by brownification are contradictory and species specific (Estlander et al. 2012, Jönsson et al. 2013, Nurminen et al. 2014). Still, via negative effects on light availability the experimentally induced brownification was likely the main mechanism for the lower prey biomass in stomachs and lower body condition in stickleback observed in this study. Mortality in YOY stickleback was overall negatively size dependent but brownification caused a higher mortality. Increased starvation mortality due to decreased foraging rates as a consequence of decreased light availability were likely the main mechanism behind the observed higher mortality with brownification. Thus, our results suggest that when fish are able to feed during winter, increased climate change induced brownification may negatively affect search efficiency and intake rates in fish which in turn may cause increased winter mortality. Moreover, benthic production is expected to decrease with increasing DOC levels, invertebrate benthic prey production will decreases (Karlsson et al. 2009, Craig et al. 2015), which is the main recourse for fish during winter (Byström et al. 2006). Thus as benthic prey abundance may decrease with brownification, this may decrease resource availability for fish during winter which then will further strengthen the observed negative effects on YOY survival observed in this study.

III: Population responses to climate change

Chapter I was a conceptual study of a system where the spawning adults had obtained their obtained their energy for both somatic growth and reproduction in another habitat. In 2013, the stickleback populations in our experimental ecosystems constituted of individuals that all had obtained their energy for both somatic growth and reproduction within in the

14

ecosystems. Thus, in this study we studied how climate change will affect fish production and biomass at the population level in lentic closed system like lakes. Both warming and increased TDOM caused substantial effects on density and biomass of fish but as resource levels were similar between treatments this indicated that observed effects on fish populations was not strongly influenced by transient dynamics, but instead reflecting reasonable stable conditions balancing resource production to consumption and growth rates of fish (Hairston et al. 1960, Carpenter et al. 1985). TDOM addition caused both the expected increased pelagic GPP and decreased benthic GPP, likely due to the increased nutrients and decreased light availability that followed with the TDOM water input (Karlsson et al., 2009, 2015). The counteracting effects in the different habitats however resulted in no effect on whole ecosystem GPP, although the TDOM increase caused a substantial reduction in fish production. Sticklebacks diets were in all treatments and to a large extent dominated by benthic resources. Hence the similar and high reliance of benthic resources in both ambient and TDOM treatments despite decreased light conditions highlights the importance of benthic production for fish growth and production even in such a small species as sticklebacks.

Whereas the effects of TDOM differed between chapter I and III due to the higher TDOM load in this study, the effects of warming at fish population level were similar in the two chapters. Primary production is expected to increase with temperature but the magnitude of increase depends on nutrient availability (O’Connor, 2009). As nutrient concentrations in ambient and warm enclosures and light availability similar and high this indicated that nutrients availability likely constrained positive responses in GPP in warm enclosures. Theoretical studies of the effects of warming on fish populations generally predict; increased maintenance requirements in fish in relation to both GPP and resource production will increase competition and cause reduced fish biomass and production and increased ratio of YOY to older fish (Ohlberger et al. 2011, Ohlberger 2013, Fussmann 2014). As whole ecosystem GPP did not increase with warming in this study, the expected increase in metabolic requirements in fish were not counteracted by any increase in primary production and likely not in resource production either. Although, cohort competition was present also in ambient control conditions, the substantial lower biomass and estimated net production of both YOY and older fish with warming suggests a strong increase in metabolic requirements in fish in relation to resource production to be the main cause of the negative effects warming on biomass and production of sticklebacks.

Despite the strong negative effects of warming and TDOM alone in this study, the expected synergic effects of the two did not decrease fish

15

production further. The combined effect was extenuating and caused less negative impact on fish production. Hence, the negative effect of increased TDOM alone on fish production was to some extent mitigated by warming or vice versa. There are several potential mechanisms that could explain the lack of negative additive effect on fish production. First of all, the additional nutrients in the TDOM water may have stimulated pelagic GPP to such extent that this counteracted the negative effect of warming alone on fish production. Secondly, DOC concentration and corresponding effects on Kd were lower in the combined treatment compared to TDOM treatment alone despite similar input of TDOM, likely as a result of temperature stimulation of heterotrophic bacterial uptake and degradation of DOC. This increased bacterial activity/production may provide an additional energy source for zooplankton and benthic invertebrates in term increasing resource production for fish (Lefébure et al., 2013). Finally, increased search efficiency due to better light conditions in comparison to TDOM treatment alone may have increased energy gain when searching for prey at low prey abundance (Estlander et al. 2012). Although neither of these effects could be could be supported with available data the they were apparently not strong enough to fully counteract the combined negative effect of warming and increased TDOM negative on fish production.

Climate change is predicted to first cause warming of aquatic ecosystems and followed by an increased TDOM as precipitations, terrestrial GPP and thawing of permafrost soils will increase (IPCC, 2013). Our results thus provide prediction of both shorter and longer term effects of climate change effects on fish populations. To summarize, the results from this study suggest that fish production will decrease in the future, first due to warming and thereafter due to input of TDOM rich water. Still, although increased resource production in pelagic systems via increased TDOM may partly counteract the negative effects of climate change on fish production, the negative effects on fish production will likely dominate and especially so in systems dominated by benthic resource production.

IV: Effects of warming on fish population demographics

Here, we studied the long term effects of warming on stickleback population dynamics and demographics. The similar lack of positive response in GPP with warming was present also in the third year (i.e. 2014) as in the previous study years (Chapter I and III). As the summer 2014 was warmer than in the two previous years the negative effects on stickleback populations should be even stronger in 2014. Correspondingly, the negative effects on stickleback

16

populations observed in 2012 and 2013 with warming was similar to the observed population responses under ambient conditions in 2014. Similarly, in warm enclosures 2014 the negative effects on stickleback population abundance were even stronger then the two previous years. Over the course of this study the temperature ranges experienced by the sticklebacks were in 2012 and 2013 close or slightly above the optimal temperature for growth i.e. 21.7 ˚C (Lefébure et al. 2011) and also still at the ascending part of the response curve for search efficiency (Lefébure et al. 2014). This suggest that the observed responses in stickleback populations were in most determined by the lack of positive responses in primary production and resulting population level feed backs on resource availability for the stickleback populations. However, the even stronger negative individual and population responses to warming observed in 2014 were likely caused at least to some extent also by physiological constraints at the individual level (Lefébure et al. 2011) during the very warmest part of this summer. Nevertheless, most of our results over the study years were in line with theoretical predictions of the effects of warming as stickleback densities and biomass decreased with warming likely as effects of decrease in ecosystem carrying capacity, an reduced energy supply for fish top consumers as a consequence of the asymmetric response to warming in energy requirements of consumers and primary production (O’Connor 2009, Ohlberger et al. 2011, Fussmann et al. 2014).

However, two responses of sticklebacks and the food webs were not in line with expected predictions of warming and both emerged as an indirect feedback effect mainly due to the strong negative effect on stickleback density with warming in 2014. First of all, over the study years the predicted increase in top down control of resources (O'Connor et al. 2009, Kratina et al. 2012, Hansson et al. 2013a) due to a temperature increase in consumption capacity at the individual changed from initial stronger to weaker compared to ambient conditions. Secondly, both YOY size and mean size at the population level increased with warming which contrasts the predicted decrease in individual size or mean size at the population level (Daufresne et al. 2009, Ohlberger et al. 2011). However, the strong negative effect on population density with warming especially in 2014 which at least can be explained by physiological temperature constraints (Lefebure et al. 2011), relaxed density dependent control of resources more in warm than in ambient enclosures. Hence, feedback mechanism at the population level may over time change initial effects or even cause the opposite predicted effects based on individual level responses to temperature. A final note can also be made based on the observed population dynamics although we only studied the population responses to warming over a three-year period. Our results on the population development showed lower variation in density over time

17

with warming which is also in line with some theoretical studies suggesting increased stability with warming (Fussmann et al. 2014, Amarasekare 2015). Although only tentative, our results point to the importance of seasonality and size dependent winter mortality as a potential mechanism promoting stability. Winter mortality was higher in ambient than in warm enclosures suggesting that winter mortality may be density dependent caused by negative density dependent effects on resource availability over winter. As winter mortality also is negatively size dependent in many fish species including sticklebacks (Byström et al. 2006, chapter II), the strong negative effect of warming on fish densities, but positive effects on individual size, may increase winter survival and thereby potentially increase densities of fish next spring to be high enough to increase likelihood for persistence. Still, the results here point to that the projected temperature increases due to ongoing climate change over time will have strong impact on fish population demographics and cause decreased fish population densities and biomass.

Concluding remarks Predictions on how populations in natural ecosystems will respond to changes in environmental factors based on result from experimental systems performed at small spatial and short temporal scales should always be treated with caution. Low population sizes may increase bias due to stochasticity, enclosure effects may mitigate or exaggerate treatment effects, difficulties to simulate the combined effects of natural climatologically influences and lack of natural population and ecosystem feed backs are all factors that increase the uncertainty in .

For example, some experimental studies predict no effects or increased fish production with warming (Hansson et al., 2013; Lefébure et al., 2013). Similarly, these studies suggest that the combined effect of warming and increasing levels of TDOM should increase fish production. However, both these studies were done in relative small mesocosms which restricted the density of fish to only a few individuals. Hence, the predicted positive effects on fish production of climate change were based on short term individual growth responses in few fish. These results should be contrasted to ours which instead suggested strong negative effects at the population level. At the same time warming still could lead to increased individual growth as warming decreased density of fish to very low levels (Chapter IV). Thus, this contrast suggests that prediction of climate change effects of production at the population level based on individual level responses should be treated with caution especially as responses are to large extent driven by ecosystem

18

feedbacks at the population levels (Woodward et al. 2010).

Although, we cannot rule out the possibility that our predictions are burdened with scale dependent errors, we argue that the experimental studies were done at a spatial scale that allowed fish population densities to be high enough to avoid stochastic factor to bias our results. Our experimental study was also performed in ecosystems with high natural resemblance including natural benthic surface sediments and natural occurring organisms at all lower trophic levels. Many experimental studies do not include a benthic habitat or natural benthic sediments (e.g. Moran et al. 2010, Kratina et al. 2012, Hansson et al. 2013, Lefebure et al. 2013) and as such they do not account for climate change effects on natural benthic primary and secondary production. Thus predictions from pelagic based studies may hold for deep aquatic ecosystems. However, the shallow coastal areas in marine systems are important for fish production (Seitz et al. 2014) and the majority of lakes in the northern hemisphere are both small and shallow (Downing et al. 2006). Thus this points to the importance of including responses in benthic habitat for accurate predictions of fish responses to climate change. Moreover, as study IV spans three seasons we allowed for natural variation in climate to influence our results but still found consistent results in line with theoretical predictions. A final and important consideration, the results in (IV) provides relative strong evidence that our methodological approach to heat the water by circulating the water through heat exchangers was not the cause behind the observed strong negative effects of warming on fish populations. In ambient non heated conditions, water temperatures in the summer 2014 were similar to temperatures in the warm treatment in 2012 and 2013. Correspondingly, the stickleback population densities and biomass decreased in ambient conditions to levels similar to what was observed with warming in 2012-13. These results thus suggest that the effect of increased water temperature per se, and not our method to heat the water, to be the cause of observed negative effect of fish populations. Hence, we believe that our results provide reasonable well founded predictions of how climate change induced warming and increased input of TDOM will affect fish populations.

19

Climate change is predicted to cause warming of +3 to 5 ˚C of aquatic ecosystems and the ongoing warming will be followed by an increased TDOM input to aquatic systems as precipitations, terrestrial GPP and thawing of permafrost soils will increase (IPCC 2013). Thus the experimental approach in this thesis can provide predictions of expected short and long term effects of climate change on fish populations. To summarize the key findings in this thesis; predicted future climate change induced warming and brownification of aquatic ecosystems are likely to cause decreased fish population densities, biomass and production.

20

References

Allen, A. P., J. F. Gillooly, and J. H. Brown. 2005. Linking the global carbon cycle to individual metabolism. Functional Ecology 19:202–213.

Amarasekare, P. 2015. Effects of temperature on consumer-resource interactions. Journal of Animal Ecology 84:665–679.

Ask, J., J. Karlsson, L. Persson, P. Ask, P. Byström, and M. Jansson. 2009a. Whole-lake estimates of carbon flux through algae and bacteria in benthic and pelagic habitats of clear-water lakes. Ecology 90:1923–1932.

Ask, J., J. Karlsson, L. Persson, P. Ask, P. Byström, and M. Jansson. 2009b. Terrestrial organic matter and light penetration: Effects on bacterial and primary production in lakes. Limnology and Oceanography 54:2034–2040.

Bartels, P., P. E. Hirsch, R. Svanbäck, and P. Eklöv. 2016. Dissolved Organic Carbon Reduces Habitat Coupling by Top Predators in Lake Ecosystems. Ecosystems 19:955–967.

Baulch, H. M., D. W. Schindler, M. A. Turner, D. L. Findlay, M. J. Paterson, and R. D. Vinebrooke. 2005. Effects of warming on benthic communities in a boreal lake: implications of climate change. Limnology and Oceanography 50:1377–1392.

Beisner, B. E., E. Mccauley, and F. J. Wrona. 1997. Predator–prey instability: individual-level mechanisms for population-level results. Functional Ecology 11:112–120.

Biro, P. A., A. E. Morton, J. R. Post, and E. A. Parkinson. 2004. Over-winter lipid depletion and mortality of age-0 rainbow trout ( Oncorhynchus mykiss ). Canadian Journal of Fisheries and Aquatic Sciences 61:1513–1519.

Botrell, H., A. Duncan, Z. Gliwicz, E. Grygierik, P. Larsson, and T. Weglenska. 1976. A review of some problems in zooplankton production studies. Norwegian Journal of Zoology 24:419–456.

Brett, J. 1979. Environmetal factors and growth. Pages 559–675 in W. S. Hoar, D. J. Randall, and J. R. Brett, editors. Fish fysiology. Academics Press, New York.

21

Byström, P., J. Andersson, A. Kiessling, and L. Eriksson. 2006. Size and temperature dependent foraging capacities and metabolism: consequences for winter starvation mortality in fish. Oikos 115:43–52.

Byström, P., L. Persson, and E. Wahlström. 1998. Competing predators and prey: juvenile bottlenecks in whole-lake experiments. Ecology 79:2153–2167.

Byström, P., L. Persson, E. Wahlström, and E. Westman. 2003. Size-and density-dependent habitat use in predators: consequences for habitat shifts in young fish. Journal of Animal Ecology 72:156–168.

Carpenter, S. R., J. F. Kitchell, and J. R. Hodgson. 1985. Cascading trophic interactions and lake productivity. BioScience 35:634–639.

Craig, N., S. E. Jones, B. C. Weidel, and C. T. Solomon. 2015. Habitat, not resource availability, limits consumer production in lake ecosystems: Habitat drives zoobenthos production. Limnology and Oceanography 60:2079–2089.

Cross, W. F., J. M. Hood, J. P. Benstead, A. D. Huryn, and D. Nelson. 2015. Interactions between temperature and nutrients across levels of ecological organization. Global Change Biology 21:1025–1040.

Diehl, S., S. Berger, and R. Wöhrl. 2005. Flexible nutrient stoichiometry mediates environmental influences on phytoplankton and its resources. Ecology 86:2931–2945.

Downing, J. A., Y. T. Prairie, J. J. Cole, C. M. Duarte, L. J. Tranvik, R. G. Striegl, W. H. McDowell, P. Kortelainen, N. F. Caraco, J. M. Melack, and others. 2006. The global abundance and size distribution of lakes, ponds, and impoundments. Limnology and Oceanography 51:2388–2397.

Dring, M. 1984. Light and Photosynthesis in Aquatic Ecosystems - Kirk,jto. Nature 309:382–382.

Dumont, H. J., I. V. de Velde, and S. Dumont. 1975. The dry weight estimate of biomass in a selection of Cladocera, Copepoda and Rotifera from the plankton, periphyton and benthos of continental waters. Oecologia 19:75–97.

Englund, G., G. Öhlund, C. L. Hein, and S. Diehl. 2011. Temperature dependence of the functional response. Ecology letters 14:914–921.

22

Eriksson Wiklund, A.-K., K. Dahlgren, and B. Sundelin. 2009. Effects of warming and shifts of pelagic food web structure on benthic productivity in a coastal marine system.pdf. Mar Ecol Prog Ser 396:13–25.

Estlander, S., J. Horppila, M. Olin, M. Vinni, H. Lehtonen, and L. Nurminen. 2012. Troubled by the humics — effects of water colour and interspecific competition on the feeding efficiency of planktivorous perch. Boreal Environment Research 2012:305–312.

Feuchtmayr, H., D. McKee, I. F. Harvey, D. Atkinson, and B. Moss. 2007. Response of macroinvertebrates to warming, nutrient addition and predation in large-scale mesocosm tanks. Hydrobiologia 584:425–432.

Fussmann, K. E., F. Schwarzmüller, U. Brose, A. Jousset, and B. C. Rall. 2014. Ecological stability in response to warming. Nature Climate Change 4:206–210.

Gross, H. P. 1978. Observations on the geographic variation of the marine coastal fish Spinachia spinachia. Marine Biology 47:297–302.

Hairston, N. G. 1993. Cause-Effect Relationships in Energy Flow, Trophic Structure, and Interspecific Interactions. The American Naturalist 142:379–411.

Hairston, N. G., F. E. Smith, and L. B. Slobodkin. 1960. Community structure, population control, and competition. American naturalist:421–425.

Hansson, L.-A., A. Nicolle, W. Granéli, P. Hallgren, E. Kritzberg, A. Persson, J. Björk, P. A. Nilsson, and C. Brönmark. 2013. Food-chain length alters community responses to global change in aquatic systems. Nature Climate Change 3:228–233.

Hecky, R. E., and R. H. Hesslein. 1995. Contributions of benthic algae tolake food webs as revealed by stable isotope analysis.pdf. Journal of the North Amerivan Benthological Society 14:631–653.

Helland, I. P., A. G. Finstad, T. Forseth, T. Hesthagen, and O. Ugedal. 2011. Ice-cover effects on competitive interactions between two fish species: Ice-cover and competition. Journal of Animal Ecology 80:539–547.

Hendersson, B. A., and E. H. Brown. 1985. Effects of abundance an water temperature on recruitment and growth af alewife (Alosa

23

pseudoharengus)near South Bay, Lake Huron, 1954-192. Canadian Journal of FIsheries and Aquatic Science:1608–1613.

Hotchkiss, E. R., and R. O. J. Hall. 2014. High rates of daytime respiration in three streams: Use of δ 18 O O2 and O 2 to model diel ecosystem metabolism. Limnology and Oceanography 59:798–810.

IPCC. 2013. Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA.

Jansson, M. 1998. Nutrient Limitation and Bacteria — Phytoplankton Interactions in Humic Lakes. Pages 177–195 in D. O. Hessen and L. J. Tranvik, editors. Aquatic Humic Substances: Ecology and Biogeochemistry. Springer Berlin Heidelberg, Berlin, Heidelberg.

Jansson, M., L. Persson, A. M. De Roos, R. I. Jones, and L. J. Tranvik. 2007. Terrestrial carbon and intraspecific size-variation shape lake ecosystems. Trends in ecology & evolution 22:316–322.

Jobling, M. 1995. Fish bioenergetics. Springer.

Jonsson, M., P. Hedström, K. Stenroth, E. R. Hotchkiss, F. R. Vasconcelos, J. Karlsson, and P. Byström. 2015. Climate change modifies the size structure of assemblages of emerging aquatic insects. Freshwater Biology 60:78–88.

Jönsson, M., S. Hylander, L. Ranåker, P. A. Nilsson, and C. Brönmark. 2011. Foraging success of juvenile pike Esox lucius depends on visual conditions and prey pigmentation. Journal of Fish Biology 79:290–297.

Jönsson, M., L. Ranåker, P. A. Nilsson, C. Brönmark, and J. Grant. 2013. Foraging efficiency and prey selectivity in a visual predator: differential effects of turbid and humic water. Canadian Journal of Fisheries and Aquatic Sciences 70:1685–1690.

Karlsson, J., A.-K. Bergström, P. Byström, C. Gudasz, P. Rodríguez, and C. Hein. 2015. Terrestrial organic matter input suppresses biomass production in lake ecosystems. Ecology 96:2870–2876.

Karlsson, J., and P. Byström. 2005. Littoral energy mobilization dominates energy supply for top consumers in subarctic lakes. Limnology and Oceanography 50:538–543.

24

Karlsson, J., P. Bystrom, J. Ask, P. Ask, L. Persson, and M. Jansson. 2009. Light limitation of nutrient-poor lake ecosystems. Nature 460:506–509.

Klug, J. L. 2005. Bacterial response to dissolved organic matter affects resource availability for algae. Canadian Journal of Fisheries and Aquatic Sciences 62:472–481.

Kratina, P., H. S. Greig, P. L. Thompson, T. S. Carvalho-Pereira, and J. B. Shurin. 2012. Warming modifies trophic cascades and eutrophication in experimental freshwater communities. Ecology 93:1421–1430.

Lefébure, R., R. Degerman, A. Andersson, S. Larsson, L.-O. Eriksson, U. Båmstedt, and P. Byström. 2013. Impacts of elevated terrestrial nutrient loads and temperature on pelagic food-web efficiency and fish production. Global Change Biology 19:1358–1372.

Lefébure, R., S. Larsson, and P. Byström. 2011. A temperature-dependent growth model for the three-spined stickleback Gasterosteus aculeatus. Journal of Fish Biology 79:1815–1827.

Lefébure, R., S. Larsson, and P. Byström. 2014. Temperature and size-dependent attack rates of the three-spined stickleback (Gasterosteus aculeatus); are sticklebacks in the Baltic Sea resource-limited? Journal of Experimental Marine Biology and Ecology 451:82–90.

López-Urrutia, Á., E. San Martin, R. P. Harris, and X. Irigoien. 2006. Scaling the metabolic balance of the oceans. Proceedings of the National Academy of Sciences 103:8739–8744.

Moran, R., I. Harvey, B. Moss, H. Feuchtmayr, K. Hatton, T. Heyes, and D. Atkinson. 2010. Influence of simulated climate change and eutrophication on three-spined stickleback populations: a large scale mesocosm experiment. Freshwater Biology 55:315–325.

Müren, U., J. Berglund, K. Samuelsson, and A. Andersson. 2005. Potential effects of elevated sea-water temperature on pelagic food webs. Hydrobiologia 545:153–166.

Murphy, J., and J. P. Riley. 1962. A modified single solution method for the determination of phosphate in natural waters. Analytica Chimica Acta 27:31–36.

Nurminen, L., S. Estlander, M. Olin, and H. Lehtonen. 2014. Feeding efficiency of planktivores under disturbance, the effect of water colour, predation threat and shoal composition: feeding of

25

planktivores under disturbance. Journal of Fish Biology 84:1195–1201.

O’Connor, M. I. 2009. Warming strengthens an herbivore-plant interaction. Ecology 90:388–398.

O’Connor, M. I., B. Gilbert, and C. J. Brown. 2011. Theoretical Predictions for How Temperature Affects the Dynamics of Interacting Herbivores and Plants. The American Naturalist 178:626–638.

Ohlberger, J. 2013. Climate warming and ectotherm body size - from individual physiology to community ecology. Functional Ecology 27:991–1001.

Ohlberger, J., E. Edeline, L. A. Vøllestad, N. C. Stenseth, and D. Claessen. 2011. Temperature-driven regime shifts in the dynamics of size-structured populations. The American Naturalist 177:211–223.

Persson, L. 1986. Effects of reduced interspecific competition on resource utilization in perch (Perca fluviatilis). Ecology:355–364.

Persson, L., J. Andersson, E. Wahlstrom, and P. Eklov. 1996. Size-Specific Interactions in Lake Systems: Predator Gape Limitation and Prey Growth Rate and Mortality. Ecology 77:900.

Persson, L., A. M. de Roos, K. Leonardsson, M. Gyllenberg, and B. Christensen. 1998. Ontogenetic Scaling of Foraging Rates and the Dynamics of a Size-Structured Consumer-Resource Model. Theoretical Population Biology 54:270–293.

Petchey, O. L., P. T. McPhearson, T. M. Casey, and P. J. Morin. 1999. Environmental warming alters food-web structure and ecosystem function. Nature 402:69–72.

Post, J. R., and D. O. Evans. 1989. Size-Dependent Overwinter Mortality of Young-of-the-Year Yellow Perch (Perca flavescens): Laboratory, In Situ Enclosure, and Field Experiments. Canadian Journal of Fisheries and Aquatic Sciences 46:1958–1968.

Rand, P. S., and D. J. Stewart. 1998. Prey fish exploitation, salmonine production, and pelagic food web efficiency in Lake Ontario. Canadian Journal of Fisheries and Aquatic Sciences 55:318–327.

Reichstein, B., L. Persson, and A. M. De Roos. 2015. Ontogenetic asymmetry modulates population biomass production and response to harvest. Nature Communications 6:6441.

26

Rodríguez, P., P. Byström, E. Geibrink, P. Hedström, F. R. Vasconcelos, and J. Karlsson. 2016. Do warming and humic river runoff alter the metabolic balance of lake ecosystems? Aquatic Sciences 78:717–725.

Seekell, D. A., J.-F. Lapierre, J. Ask, A.-K. Bergström, A. Deininger, P. Rodríguez, and J. Karlsson. 2015. The influence of dissolved organic carbon on primary production in northern lakes: Influence of DOC on production. Limnology and Oceanography 60:1276–1285.

Seitz, R. D., H. Wennhage, U. Bergstrom, R. N. Lipcius, and T. Ysebaert. 2014. Ecological value of coastal habitats for commercially and ecologically important species. ICES Journal of Marine Science 71:648–665.

Sibly, R. M., J. H. Brown, and A. Kodric-Brown. 2012. Metabolic Ecology: A Scaling Approach. John Wiley & Sons.

Šorf, M., T. A. Davidson, S. Brucet, R. F. Menezes, M. Søndergaard, T. L. Lauridsen, F. Landkildehus, L. Liboriussen, and E. Jeppesen. 2015. Zooplankton response to climate warming: a mesocosm experiment at contrasting temperatures and nutrient levels. Hydrobiologia 742:185–203.

Sponaugle, S., K. Grorud-Colvert, and D. Pinkard. 2006. Temperature-mediated variation in early life history traits and recruitment success of the coral reef fish Thalassoma bifasciatum in the Florida Keys. Marine Ecology Progress Series 308:1–15.

Sterner, R. W., J. J. Elser, E. J. Fee, S. J. Guildford, and T. H. Chrzanowski. 1997. The light: nutrient ratio in lakes: the balance of energy and materials affects ecosystem structure and process. The American Naturalist 150:663–684.

Strecker, A. L., T. P. Cobb, and R. D. Vinebrooke. 2004. Effects of experimental greenhouse warming on phytoplankton and zooplankton communities in fishless alpine ponds. Limnology and Oceanography 49:1182–1190.

Sundblad, G., U. Bergstrom, A. Sandstrom, and P. Eklov. 2014. Nursery habitat availability limits adult stock sizes of predatory coastal fish. ICES Journal of Marine Science 71:672–680.

Tixier, G., K. P. Wilson, and D. D. Williams. 2008. Exploration of the influence of global warming on the chironomid community in a manipulated shallow groundwater system. Hydrobiologia 624:13–27.

27

Toneys, M. L., and D. W. Coble. 1979. Size-Related, First Winter Mortality of Freshwater Fishes. Transactions of the American Fisheries Society.

Tranvik, L. J. 1988. Availability of dissolved organic carbon for planktonic bacteria in oligotrophic lakes of differing humic content. 16:311–322.

Ulvan, E. M., A. G. Finstad, O. Ugedal, and O. K. Berg. 2012. Direct and indirect climatic drivers of biotic interactions: ice-cover and carbon runoff shaping Arctic char Salvelinus alpinus and brown trout Salmo trutta competitive asymmetries. Oecologia 168:277–287.

Vasseur, D. A., and K. S. McCann. 2005. A Mechanistic Approach for Modeling Temperature-Dependent Consumer-Resource Dynamics. The American Naturalist 166.

Vogel, J. L., and D. A. Beauchamp. 1999. Effects of light, prey size, and turbidity on reaction distances of lake trout (Salvelinus namaycush) to salmonid prey. Canadian Journal of Fisheries and Aquatic Sciences 56:1293–1297.

Vucic-Pestic, O., R. B. Ehnes, B. C. Rall, and U. Brose. 2011. Warming up the system: higher predator feeding rates but lower energetic efficiencies: WARMING AND FUNCTIONAL RESPONSES. Global Change Biology 17:1301–1310.

Woodward, G., D. M. Perkins, and L. E. Brown. 2010. Climate change and freshwater ecosystems: impacts across multiple levels of organization. Philosophical Transactions of the Royal Society B: Biological Sciences 365:2093–2106.

Wootton, R. J. 1973. Fecundity of the three-spined stickleback, Gasterosteus aculeatus (L.). Journal of fish biology.

Wootton, R. J. 1984. A functional biology of sticklebacks. University of California Press.

Yvon-Durocher, G., J. I. Jones, M. Trimmer, G. Woodward, and J. M. Montoya. 2010. Warming alters the metabolic balance of ecosystems. Philosophical Transactions of the Royal Society B: Biological Sciences 365:2117–2126.

Zippin, C. 1956. An Evaluation of the Removal Method of Estimating Animal Populations. Biometrics 12:163.

28

Tack

Då var den här resan slut. Japp. Jag vet det är en klyscha men den funkar bra. Eller är det dags ett nytt resmål? Hittills har färden pågått i 42 år. Och de senaste fem åren har jag haft förmånen att resa i ett mycket angenämt, trevligt och glatt sällskap. Ett glatt sällskap som alltid funnits tillhands både i med och motgång, glädje och frustration (här kan källorna vara många men att publicera går inte bara randrätt som man säger i textilbranchen). Men jag är inte bitter om ni tillåter en stulen fras. Med mitt sällskap kan man då rycka i tofsen och genast kommer en tänkare till och delar grubbel och tvivel på världens ordning. Med ny ork och kraft med gemensam ansträngning har kolossen rört sig framåt. Eller koloss, en avhandling i alla fall. Det har känts som en koloss i stunder av missmod, orubblig och omöjlig att rubba minsta bit framåt. I stunder av självinsikt har även jag kunnat se att kolossen /avhandlingen är en liten del för världen men en rätt stor del av min vakna och sovande tillvaro. Och nu är den klar. Men så klart med allas hjälp. För annars hade det inte blivit nånting.

Så därför vill jag säga tack till er mina vänner, kollegor och andra medresenärer som varit del av den här resan under så många år.

Tack Olle Söderström för att du öppnade det stora biologiskåpet i mitten av 90-talet. Det var aldrig tänkt att jag skulle intressera mig för biologi - fysik var ju mycket roligare. Åsså gick det att räkna precis med fysik. Skönt. Så kommer Olle och hela gänget från biologiundervisningen för lärare under sen 1990-tal och förstör allt ihop. Livet som finns där ute var faktiskt intressant. Tack för det Olle och resten!

Tack Birgitta Nordin. För all ordning på andra som du kunnat bjuda på. Birgitta var en fast punkt i tillvaron när all bara var en endas stor oreda. Alltid vänlig och enkel att fråga. Tack Birgitta!

Tack Göran för telefonsamtalet i mars 2009. Utan det hade det nog aldrig blivit av!

Tack Henrik Jensen för att du tipsade Göran. Och en massa annat såklart såsom jakt och fiske och ditt alltid lika självklart generösa sätt.

Gunnar. Tack för sällskapet under den tid vi rest tillsammans. Tack för grundkurser och mer avancerade i gäddans filosofi och psykologi. Skidåkning eller mest prat därom kanske, men i alla fall. Åka gjorde vi aldrig

29

tillsammans, men mycket prat om… Och tiden är faktiskt inte slut ännu. Tack för din tid fast du inte hade egen.

Henrik S. Tack för ditt aldrig sinande ödmjuka tålamod med en dyskalkylektiker som jag: Och att du lurade ut mig i matematiken förunderliga vatten. Men nånstans blev det ändå för djupt…

Stickan. Tack för ditt oerhörda engagemang för mångfald, mot enfald, för rättvisa och ändlösa förråd av artkunskap och exkursionsanekdoter. Det har varit ett sant nöje och sannerligen oerhört nyttigt att få hänga med på dina meandringar i skog och mark. Tack Stickan!!

Anders Nilsson. Tack för din öppna famn i systematikens namn. Din syn på klassrummets pedagogik.

Mariana och Carola. För att ni alltid lyckades finnas till hands när det brann i knutarna för att nån (det behöver inte vara jag) inte beställt eller helt enkelt tänkt fel. Att ni alltid var beredda att hjälpa till.

Ett stort tack till mitt eget R-Core-dreamteam. Ett sånt skulle alla ha. David, Gustav, Tom och Caroline. Kraftigt blottryckssänkande effekter!

Statistics Dreamteam – Tom, Danny, Johan O, Gustav. Hur skulle detta slutat utan er?! Heteroskedastiskt överspridda, utan log-lik-ratios med faktoriell tidsanalys utan slope i randomvariabeln?! Eller räcker det med en enkel chi-fick-du?! Framför mer pragmatik i statistiken tack! Men att jag lärt mig mycket genom alla vindlingar genom statistikens meandringar det torde stå utom allt rimligt tvivel.

Erik. För sällskap i vått och torrt, i fritid som arbete, jakt, fiske, cykel, skidor, squash, byggnationer, allehanda tekniska lösningar, ingenjörskonst, fria tankar, etc. Och inte minst - låt för alltid gubben i dig blomma!

Micke P. För ditt mod att skratta åt (inte med, tro inte det) doktorander som håller på att rämna av frustration. För ditt mod att möta nya naturupplevelser. Jerringprisets bragdmedalj för dina uppoffringar till fjälls på turskidor!! Miltals enkel väg för att släpa skoter genom lårdjupt flödvatten i beckmörker, för att dan därpå komma på att man dessutom hatar skidåkning kryddat med noll ripor. Tappert Micke! Vi gör om det!

30

Tack Johan förberedvillgheten! Jenn sett e et fast! Ållte leka råolit fåra dell skåogs vä dä!

Mina squashantagonister genom tiderna. Carsten, Erik, Pia Mårten, Micke (du skulle inte gett upp så tidigt i din karriär), Sven, David (alltid kul med nya angreppssätt).

Skid- och orienteringsklubben för alla skidsamtal och teknikkurser. Tack för det Åsa, Håkan, Christian Johan och Tom!

Internationella klubben! Pia, Carolina, Wojciech, Carsten, Martina, Sveti, Dagmar, Francisco, mina världsmedborgare. Vilken rikedom ni bidrar med som kommer till landet i norr och delar med er av era kulturella skatter och gör min värld aningen större.

Sven. För meandringar i musik, filosofi, politik. Ja vad allt har vi inte hunnit prata om under våra stunder. Och ett stort tack för spiggen och båttuskorven som pryder framsidan - Leve Salvadors konstinfluenser!

Tack Tomas och Micke. För att ni alltid finns och bryr er om andras vetenskapliga grubblerier. Ni skapar verkligen trevnad och psykosocial arbetsmiljö av det goda slaget! Tack!

Janne. Att alltid ha tid och möta bekymrade doktorander med ett intresse och leende. Starkt jobbat. Tack! Med er i sällskap tror jag man kan göra vilka resor som helst.

Pelle. Tack för tiden och platsen, uppmärksamhet. Uppmuntran och flatgarv när doktoranden tycks ha svåra stunder. Flatgarv som får svärtan att skimra i nyanser av möjlighet. Jag bara fattar inte hur du hinner.

Vilket radarpar ni varit under mina år som doktorand. Jag förstår inte riktigt var ni får tiden och orken. Visionära och med fötterna på jorden, klarsynta och ödmjuka. Vilken lycka att få träffa er!!

Till slut vill jag även tacka min familj. Det är ändå ni som betyder allt. Hela vägen gick ni med mig!

Nu kör vi!