Evidence for non-random spatial positioning of migrating smolts (Salmonidae) in a small lowland stream

1. The ontogenetic development of anadromous salmonids includes downstream emigration of immature individuals from freshwater towards the marine environment. Although this migration of juvenile salmonids (smolts) may be associated with severe mortalities, only limited attention has been paid to the spatial positioning of smolts in small streams. 2. Using a novel approach, this study examined the vertical and horizontal positioning of brown trout and Atlantic salmon smolts while performing downstream migration in a small lowland stream. 3. Pre‐smolts of indigenous and hatchery‐reared (F1) brown trout (Salmo trutta), and two different populations of Atlantic salmon (S. salar), were tagged with passive integrated transponder (PIT) tags and subsequently released upstream of an antenna array consisting of five circular swim‐through PIT antennas. Antennas were positioned in order to determine whether the migrating smolts were bottom or surface oriented, and if they were oriented towards the mid‐channel or the stream bank. 4. During the smolt emigration period, data describing both the detection of the migrating fish and the amount of water passing through the antennas were collected. This was accomplished in order to determine if the fish were performing active positioning behaviour independently of the vertical and horizontal discharge distributions in the stream. 5. The results showed that the smolts migrated in a non‐random spatial pattern independently of the stream discharge distributions. Vertically, the indigenous brown trout and the Atlantic salmon demonstrated a preference for the bottom orientated positions. In contrast, the distribution of the F1 brown trout was not different from the discharge distribution. The latter observation suggests random vertical positioning, which may be indicative of inferior migratory performance. Horizontally, all tested smolt populations strongly preferred the mid‐channel positions. 6. The discharge‐corrected preferences for certain spatial positions suggest that smolt emigration is not entirely a matter of passive displacement in lowland streams. 7. Anthropogenically altered channels may inhibit or delay downstream emigration of smolts resulting in increased mortalities. Given that the smolts in this study actively selected spatial positions in the mid‐channel and near the bottom, it is suggested that deep, mid‐channel furrows may be used to help guide migrating smolts past adverse habitats in lowland streams.