Emil Kristensen:
Freshwater fish in new lakes. Colonization, Species richness, Management, and Environmental factors

Date: 16-07-2020    Supervisor: Kaj Sand-Jensen

Land-use changes have reduced the number of shallow lakes and wetlands throughout both North America and Europe between 1750 and 1980, where up to 80% have been drained and in many instances claimed for agricultural land. To counter this development, the establishment of new lakes has gradually gained momentum throughout Europe and North America during the last 40 years as a way to remove nutrients, manage extreme rain events, establish recreational areas, and increase biodiversity. Globally, more than 184,000 square kilometers of new permanent water bodies have appeared between 1984 and 2015, mostly due to the construction of reservoirs. In Denmark alone, more than 70 lakes over 10 ha have been established during the preceding 40 years with many more planned for establishment. An important species group in lakes are freshwater fish which to a large degree influence trophic interactions. In this thesis, I investigate how the colonization of fish takes place, the determinants for fish species richness, management possibilities, and potential pitfalls for the fish communities in newly created lakes.

Concerning colonization of the new lakes, we found that fish migrated quickly into a newly established lake from nearby refugia with fast initial colonization followed by saturation of the species pool. The fish community composition showed increasing similarity over time compared to that in natural lakes in the region and even more to the two directly connected lakes regarding both species occurrence, catch per unit of effort (CPUE), and biomass per unit of effort (BPUE). On a broader scale, we found that species richness was not different in new lakes compared to natural lakes if they were connected to the stream network. However, the nonconnected new lakes had significantly fewer species than the natural lakes.

The species in 892 drainage basins, 179 natural lakes, and 52 new lakes, consisted most often of perch (Perca fluviatilis), roach (Rutilus rutilus), and pike (Esox lucius) and we generally found the same species in natural and new lakes. However, some species differed between the two lake types. Species such as ruffe (Gymnocephalus cernua), bream (Abramis brama), and white bream (Blicca bjoerkna), were in fewer of the new lakes compared to the natural lakes. However, species like tench (Tinca tinca), crusian carp (Carassius carassius), and common carp (Cyprinus carpio), were found more often in new lakes than in natural lakes. Following the immigration pattern in a single new lake, we found that the first colonizer was three-spined stickleback (Gasterosteus aculeatus) followed by roach and pike, which dominated the fish biomass in the following years. However, perch immigrated in low numbers and showed, together with pike, limited recruitment success. This could prove problematic since large predatory fish species are important for the trophic structure in lakes.

Management and monitoring of large predatory fish are, however, difficult to perform without stressing or even killing the individuals. I tested the ability of a non-invasive image recognition program to identify individual recaptured pike based on photos of their skin pattern. We and local anglers caught 209 pike, including 45 recaptures, over the course of 1.5 years. The software recognized all 45 recaptures which enabled us to predict a population size of 560 pike in the lake and reflect a much more comprehensive size structure than seen using standardized gillnet sampling which only yielded 4 individuals. Angling and photographic identification, involving reporting by citizen-scientists, provided valuable temporal data with multiple benefits compared to traditional sampling. However, while a healthy population of predatory fish has a positive influence on the ecological quality of the lake, high nutrients loadings offer an often-larger negative impact.

New lakes established on former fertilized agricultural fields can encounter extensive nutrient release from flooded soils that can maintain a poor ecological quality. The extent of the period with high sediment nutrient release is poorly known due to few detailed lake restoration studies. The nutrient release can have a negative effect on the fish composition increasing the proportion of zooplantivorous fish biomass compared to piscivorous fish biomass. We found that the sediment density in shallow Lake Fil was highly positively correlated with the amount of organic phosphorus in the sediment. Using sonar scanning to determined sediment density over the entire lake bottom, we calculated that 65 tons P of the initial 163 tons P in the top 9 cm of the sediment had been lost from the newly established lake between October 2012 and September 2015. The phosphorus loss can be referred to high wind exposure of the sediments in the shallow lake waters, and short water retention time leading to sediment particle resuspension and washout. High sediment exposure and higher water renewal could be a useful combination in future lake creation projects on former agricultural land to enable quick removal of phosphorus from the lake. However, creating lakes in an agricultural landscape can potentially introduce other problems for the fish communities.

We observed a new climate-related cause of fish kills in the re-established shallow Lake Fil. A long period of drought in a hot summer followed by heavy rain resulted in a large input of labile organic matter to the lake. This was followed by a condition of whole-lake anoxia and fish kill in the lake basin receiving the input, while the second basin, located immediately downstream, was left unaffected. An oxygen model suggested that respiration had increased by about 230%, following the organic input, and caused whole-lake nocturnal anoxia for four days despite unaltered daytime photosynthesis. One year after the fish kill, we observed a large change in the fish composition; roach and bream were high in numbers while large predatory perch and pike remained very few. This imbalance in the fish food webs may last for several years and can in turn increase predation on zooplankton and release phytoplankton from grazing control. The prolonged effects of a fish kill on fish and lake community structure demand further research, as weather-induced anoxia can be expected to become more common. Especially in shallow lakes, natural or created, in agricultural well-drained areas.