Anders Hargeby

Trophic web structure in a shallow eutrophic lake during a dominance shift from phytoplankton to submerged macrophytes

In Lake Krankesjön, southern Sweden, sago pondweed (Potamogeton pectinatus L.) and a stonewort (Chara tomentosa L.) expanded spatially during the second half of the 1980s after more than a decade of phytoplankton blooms and sparse submerged vegetation. During the expansion of submerged plants the number of resting and breeding waterfowl increased. The increase was significant for herbivorous birds such as coot (Fulica atra L.) and mute swan (Cygnus olor (Gmelin)), but also for omnivorous dabbling ducks. The shift from phytoplankton to submerged macrophytes caused structural changes on higher trophic levels, and an altered trophic web developed. The density of planktonic Cladocera decreased, which is suggested to be a result of decreased phytoplankton productivity and biomass as nutrient levels dropped. The benthic macroinvertebrate assemblage changed from low diversity and biomass dominated by Chironomidae and Oligochaeta on bare sediment, to high diversity and biomass characterized by plant-associated forms like snails and isopods in areas covered by macrovegetation. The mean size of perch (Perca fluviatilis L.) increased, probably as a result of higher availability of macroinvertebrates in the vegetation. The perch reached a mean size where the species is known to shift to a fish diet, permitting an increased top down effect on the ecosystem. The results support the idea that shallow eutrophic lakes can shift between two states, each one stabilized by feed-back mechanisms including both biotic and abiotic factors. Shifts between these states are suggested to be a possible explanation for observed drastic changes in abundance of waterfowl in shallow eutrophic lakes.

Long-term patterns of shifts between clear and turbid states in Lake Krankesjon and Lake Takern

During the past century, Lake Tåkern and Lake Krankesjön, southern Sweden, have shifted repeatedly between a state of clear water and abundant submerged vegetation, and a state of turbid water and sparse vegetation. Long-term empirical data on such apparently alternative stable state dynamics are valuable as complements to modeling and experiments, although the causal mechanisms behind shifts are often difficult to identify in hindsight. Here, we summarize previous studies and discuss possible mechanisms behind the shifts. The most detailed information comes from monitoring of two recent shifts, one in each lake. In the 1980s, L. Krankesjön shifted to clear water following an expansion of sago pondweed, Potamogeton pectinatus. Water clarity increased when the pondweed was replaced by characeans. Zooplankton biomass in summer declined and the concentration of total phosphorus (TP) was reduced to half the previous level. The fish community changed over several years, including an increasing recruitment of piscivorous perch (Perca fluviatilis). An opposite directed shift to turbid water occurred in Lake Tåkern in 1995, when biomass of phytoplankton increased in spring, at the expense of submerged vegetation. Consistent with the findings in L. Krankesjön, phyto- and zooplankton biomass increased and the average concentration of TP doubled. After the shift to clear water in L. Krankesjön, TP concentration has increased during the latest decade, supporting the idea that accumulation of nutrients may lead to a long-term destabilization of the clear water state. In L. Tåkern, data on TP are inconclusive, but organic nitrogen concentrations oscillated during a 25-year period of clear water. These observations indicate that intrinsic processes cause gradual or periodic changes in system stability, although we cannot exclude the possibility that external forces are also involved. During such phases of destabilization of the clear water state, even small disturbances could possibly trigger a shift, which may explain why causes behind shifts are hard to identify even when they occur during periods of extensive monitoring.

Habitat-specific pigmentation in a freshwater isopod : Adaptive evolution over a small spatiotemporal scale

Abstract Pigmentation in the freshwater isopod Asellus aquaticus (Crustacea) differed between habitats in two Swedish lakes. In both lakes, isopods had lighter pigmentation in stands of submerged vegetation, consisting of stoneworts (Chara spp.), than in nearby stands of reed (Phragmites australis). Experimental crossings of light and dark isopods in a common environment showed that pigmentation had a genetic basis and that genetic variance was additive. Environmental effects of diet or chromatophore adjustment to the background had minor influence on pigmentation, as shown by laboratory rearing of isopods on stonewort or reed substrates, as well as analyses of stable isotope ratios for isopods collected in the field. In both study lakes, the average phenotype became lighter with time (across generations) in recently established stonewort stands. Taken together, these results indicate that altered phenotype pigmentation result from evolutionary responses to local differences in natural selection. Based on the assumption of two generations per year, the evolutionary rate of change in pigmentation was 0.08 standard deviations per generation (haldanes) over 20 generations in one lake and 0.22 haldanes over two generations in the other lake. This genetic change occurred during an episode of population growth in a novel habitat, a situation known to promote adaptive evolution. In addition, stonewort stands constitute large and persistent patches, characteristics that tend to preserve local adaptations produced by natural selection. Results from studies on selective forces behind the adaptive divergence suggest that selective predation from visually oriented predators is a possible selective agent. We found no indications of phenotype‐specific movements between habitats. Mating within stonewort stands was random with respect to pigmentation, but on a whole‐lake scale it is likely that mating is assortative, as a result of local differences in phenotype distribution.

Macrophyte Associated Invertebrates and the Effect of Habitat Permanence

Invertebrate abundance differed in stands of the two characeans Chara tomentosa L. and Nitellopsis obtusa (Desv.) J. Groves in a shallow eutrophic lake. The plants are similar in morphology but differ in growth cycle. In stands of the winter-green Chara tomentosa, Asellus aquaticus L. (Isopoda) dominated in abundance over Chironomidae (Diptera) and Gammarus lacustris G. O. Sars (Amphipoda) occurred frequently. In stands of Nitellopsis obtusa, which dies off during winter, Chironomidae was the dominating taxon, while the abundance of Asellus was one to two orders of magnitude lower than in Chara, and Gammarus was not found. Field and laboratory preference tests showed that Asellus and Gammarus did not colonize or prefer Chara over Nitellopsis, suggesting that active avoidance was not the cause of their low abundance in Nitellopsis. Exclusion of fish for 90 d in a Nitellopsis stand did not increase the density of Asellus, but introduction of Asellus to exclosures showed that the species was able to grow rapidly and reproduce in the habitat. In a recently established Chara stand the species structure of associated invertebrates shifted from a dominance of Chironomidae in September of the first year to a dominance of Asellus the following year. These results all support the hypothesis that the die-off of Nitellopsis in autumn limits slow colonizers like Asellus and Gammarus from establishing dense populations in this habitat. The invertebrate community in this seasonally changing habitat is thus suggested to be in an early phase of colonization, where animal interactions are less important and the species composition is mainly determined by colonization ability. In the more permanent Chara habitat, density-dependent animal interactions are suggested to determine the abundance and species composition.

Shifts between clear and turbid states in a shallow lake: multi-causal stress from climate, nutrients and biotic interactions

We used long-term monitoring data to assess causes behind a recent shift front a clear to a turbid water state in Lake Takern, Sweden. The lake has a previous record of shifts between clear-water and turbid states. but the causes behind these shifts are not well known. During the recent shift, which occurred in 1995-1997, Submerged vegetation Subsequently declined after a 30-year period of clear-water and abundant vegetation. Among the possible explanations we identify several processes unlikely to have contributed to the recent shift from clear to turbid conditions. including long-term changes in external input of phosphorus. fluctuations in water level. and changes in zooplankton grazing pressure. Instead, likely scenarios to have contributed to the macrophyte decline, and thereby to the shift were: (1) a series of mild winters with short ice cover and absence of winter-kills of fish, leading to high biomasses of benthivorous and planktivorous fish before the shift, and thereby increased bioturbation and internal nutrient recycling, (2) unusually cool and windy springs the years before and during, the shift, leading to unfavourable conditions during the establishing phase of submerged macrophytes. Both shorter periods of ice cover and high wind velocity in winter and spring were associated with climate. approximated by the North Atlantic Oscillation (NAO). We argue that none of these processes alone can force the lake front the clear to the turbid state, but that several stress factors in concert are necessary to initiate a shift. Hence, we conclude that climate variability is likely to have contributed to a multi-causal stress. reducing the resilience of the clear-water state and finally triggering the Shift through inter-year dependent changes; in biomass of submerged macrophytes and fish, organism groups known to have key roles in the dynamics of shallow lakes. (Less)

Parallelism and historical contingency during rapid ecotype divergence in an isopod

Recent studies on parallel evolution have focused on the relative role of selection and historical contingency during adaptive divergence. Here, we study geographically separate and genetically independent lake populations of a freshwater isopod (Asellus aquaticus) in southern Sweden. In two of these lakes, a novel habitat was rapidly colonized by isopods from a source habitat. Rapid phenotypic changes in pigmentation, size and sexual behaviour have occurred, presumably in response to different predatory regimes. We partitioned the phenotypic variation arising from habitat (‘selection’: 81–94%), lake (‘history’: 0.1–6%) and lake × habitat interaction (‘unique diversification’: 0.4–13%) for several traits. There was a limited role for historical contingency but a strong signature of selection. We also found higher phenotypic variation in the source populations. Phenotype sorting during colonization and strong divergent selection might have contributed to these rapid changes. Consequently, phenotypic divergence was only weakly influenced by historical contingency.

Locally differentiated cryptic pigmentation in the freshwater isopod Asellus aquaticus

A repeated pattern of background colour matching in animals is an indication that pigmentation may be cryptic. Here, we examine the relationship between pigmentation of the freshwater isopod Asellus aquaticus and background darkness in 29 lakes, wetlands and ponds in Southern Sweden. The results show that Asellus pigmentation was correlated with substrate darkness across all localities. In seven localities, in which two contrasting substrate types were noted, Asellus populations were differentiated with respect to pigmentation. These findings thus provide phenomenological support for cryptic pigmentation in Asellus. Pigmentation generally increased with body size, but the relationship between pigmentation and size differed among localities, possibly as a result of differences in correlational selection on pigmentation and size. Selection thus appears to have resulted in local differentiation over a small spatial scale, even within lakes and wetlands. This differentiation is a likely cause behind elevated phenotype variation noted in localities with two substrate types, suggesting that habitat heterogeneity promotes genetic diversity.

Alternative Stable States in Shallow Lakes: What Causes a Shift?

Lake TE5;kern and Lake Krankesjon, two shallow, moderately eutrophic, calcium-rich lakes in southern Sweden have shifted between turbid and clearwater states several times during the past decades (Fig. 26.1). Lake Krankesjon shifted from a clearwater state with abundant submerged vegetation to a turbid state with sparse vegetation during the mid-1970s (Karlsson et al., 1976) and back to a clearwater state during 1985. Today, the lake is in the clearwater state, with abundant submerged vegetation dominated by Charophyta (Blindow et al., 1993; Fig. 26.2). Both shifts coincided with deviations from the average water level. During the mid-1970s, the water level during spring and summer was about 15 cm higher than average, whereas it was about 10 cm lower than average during 1983-1985 (Blindow, 1992).

Rapid adaptive divergence between ecotypes of an aquatic isopod inferred from F-Q analysis.

Divergent natural selection is often thought to be the principal factor driving phenotypic differentiation between populations. We studied two ecotypes of the aquatic isopod Asellus aquaticus which have diverged in parallel in several Swedish lakes. In these lakes, isopods from reed belts along the shores colonized new stonewort stands in the centre of the lakes and rapid phenotypic changes in size and pigmentation followed after colonization. We investigated if selection was likely to be responsible for these observed phenotypic changes using indirect inferences of selection (FST–QST analysis). Average QST for seven quantitative traits were higher than the average FST between ecotypes for putatively neutral markers (AFLPs). This suggests that divergent natural selection has played an important role during this rapid diversification. In contrast, the average QST between the different reed ecotype populations was not significantly different from the mean FST. Genetic drift could therefore not be excluded as an explanation for the minor differences between allopatric populations inhabiting the same source habitat. We complemented this traditional FST–QST approach by comparing the FST distributions across all loci (n = 67–71) with the QST for each of the seven traits. This analysis revealed that pigmentation traits had diverged to a greater extent and at higher evolutionary rates than size‐related morphological traits. In conclusion, this extended and detailed type of FST–QST analysis provides a powerful method to infer adaptive phenotypic divergence between populations. However, indirect inferences about the operation of divergent selection should be analyzed on a per‐trait basis and complemented with detailed ecological information.

Effects of pH, humic substances and animal interactions on survival and physiological status of Asellus aquaticus L. and Gammarus pulex (L.)

SummaryThe mortality and physiological status (body water content) of Asellus aquaticus (Isopoda) and Gammarus pulex (Amphipoda) were measured after 25 days exposure in 20 natural streams with a pH range of 4.3–7.5 and a colour range of 8–280 mg Pt L−1. In addition, the effects of keeping the animals as single species or together were studied. The response of Gammarus to low pH was an increased mortality and lower physiological status of surviving individuals in streams with a pH lower than 6.0. In Asellus the physiological status was correlated with pH, while the mortality was not pH dependent. The effects of humus on the physiological status of Asellus was significant when fitted to a second order polynomial function. The influence of humus can, however, be regarded as small relative to pH. The interactions between the species could be described as asymmetric under optimal conditions of high pH and low humus concentrations, where the presence of Gammarus decreased the survival and physiological status of Asellus. Acid stress did not seem to reverse the direction of this asymmetry, but the presence of Gammarus improved the physiological status of Asellus at pH lower than 6.0. Since the presence of Asellus did not increase the mortality or decrease the physiological status of Gammarus, this could be explained by Asellus feeding on Gammarus that died from physiological stress solely. This mechanism suggests that food quality, and thus effects of diffuse competition, can be important for the ability to withstand acid stress. The results, though, give no support for the hypothesis that competition from Asellus is important for the disappearance of Gammarus during the acidification of streams.