A proportion of the smolts emigrating from the river in spring were captured in a fish trap situated in a diversionary lade (~500 m upstream from the tidal limit) as part of a programme to estimate total smolt production. The trap was operated throughout the smolt run between late March and early July each year. The lade diverted a proportion of total emigrant smolts (typically 5% - 10% depending on flow conditions) and the trap was considered to be 100% efficient within the lade except in rare cases of spillage. In autumn 2013, damage to the dyke which diverted water into the lade caused trapping at this facility to be stopped. Weekly counts of smolts at the trap collected between 1975 and 2013 are considered to be robust, with similar levels of effort, and protocols, employed throughout the time series. However, note that the proportion of smolts captured was negatively related to river flow within the season; as this was not corrected for in the data, the weekly counts do not equate to total smolt production.