Ecological theory postulates that size and isolation of habitat patches impact the colonization/extinction dynamics that determine community species richness and population persistence. Given the key role of lotic habitats for life history completion in rheophilic fish, evaluating how the distribution of swift-flowing habitats affects the abundance and dynamics of subpopulations is essential. Using extensive electrofishing data, we show that merging island biogeography with meta-population theory, where lotic habitats are considered as islands in a lentic matrix, can explain spatiotemporal variation in occurrence and density of brown trout (Salmo trutta). Subpopulations in larger and less isolated habitat patches had higher average densities and smaller between-year density fluctuations. Larger habitat patches also had lower predicted risk of excessive zero catches, indicative of lower extinction risk. Trout density further increased with distance from the edge of adjacent lentic habitats with predator (Esox lucius) presence, suggesting that edge- and matrix-related mortality contributes to the observed patterns. These results can help reduce the negative impacts that habitat loss and fragmentation have on biodiversity, by stressing the importance of suitable habitat size and connectivity, and aid in prioritization of habitat restoration, dam removal and reintroduction programs aimed at vitalizing declining and locally extinct riverine fish populations.

Data on fish presence and density was generated through electrofishing, a non-lethal fish sampling method mainly conducted in streams where it is possible to wade. Electric current (DC) is used to attract fish to swim towards a hand-held anode where they are caught with a dipping net. It is an established and reliable method for quantifying fish density. See Bohlin et al (1989) and the Swedish and European Standard (SIS. SS‐EN 14011:2003: Water quality – Sampling of fish with electricity. Stockholm: Swedish Standard Institute; 2003.) for detailed descriptions on the method.

Bohlin, T., Hamrin, S., Heggberget, T. G., Rasmussen, G., & Saltveit, S. J. (1989). Electrofishing - Theory and practice with special emphasis on salmonids. Hydrobiologia, 173(1), 9–43. https://doi.org/10.1007/BF00008596

Variables used
y0plus: density of young-of-the-year (0+) brown trout expressed as round(log10(x+1)*100) where x is individuals per 100 m².
y1plus: density of older than young-of-the-year (1+) brown trout expressed as round(log10(x+1)*100) where x is individuals per 100 m².
y_total: density of brown trout of all age classes expressed as round(log10(x+1)*100) where x is individuals per 100 m².
island_size: length of lotic stretch in meters
lg_size: as above but in natural log
dist_next_island: distance to next neighbouring lotic island in meters
lg_dist: as above but in natural log
Upstream: dichotomous variable (1/0) stating whether the next neighbouring lotic island is located upstream (1) or downstream (0)
dist_to_edge_min: distance to the edge between lotic and lentic habitat in meters 
lg_edge_dist: as above but in natural log
pike01: dichotomous variable (1/0) stating whether pike has been found on the site at any electrofishing occasion
IBG: habitat category (A, B, C, D) as per Fig 1 in the manuscript
size_of_next_lot: the length of the next neighbouring lotic stretch in meters
lg_next_size: as above but in natural log

Two random effects
RiverIdent: River identity
LoticIdent: A integerised identifier for each unique lotic stretch used as random effect

Data files:
data_trout.csv with aggregated mean densities per site
data_trout01.csv unaggregated densities per site for calculation of variability over time

Special in data_trout01:
fiskedat: date in eight numbers (YYYYMMDD)
Öring0: density of young-of-the-year (0+) brown trout in individuals per 100 m²
Öring: density of older than young-of-the-year (1+) brown trout in individuals per 100 m²
ÖringTOT: density of brown trout of all age classes in individuals per 100 m²