Anthropogenic use of water systems may cause temperature fluctuations between tributaries and large rivers for which physiological population related-effects on osmoregulatory capacity of Atlantic salmon are not well described. We simulated the downstream route in the case of the River Meuse basin to investigate the impact of a 5 °C temperature shift during smoltification on hypo-osmoregulatory capacities of smolts. Three temperature regimes were tested: control temperature-treatment (T1) without temperature shift, early (T2) or late (T3) temperature shift-treatment. Moreover, fish were subjected to seawater challenge during and after the downstream migration peak time. Two allochtonous strains were used: Loire-Allier (LA) and Cong (CG). Without temperature shift (T1), significant differences between the strains were noticed in the peak date and maximum activity of gill Na+/K+ATPase as well as in plasma sodium and potassium concentrations. For early (T2) and late (T3) temperature shift-treatments, gill Na+/K+ATPase activity, plasma osmolality and ion concentrations were negatively influenced in both strains. After salinity challenge, the highest osmolality was measured in smolts subjected to the temperature shift. Predictably circulating levels of GH and IGF-1 changed over the smolting period but they did not explain the observed modifications in hypo-osmoregulatory abilities whatever the population. The results show a negative impact of a temperature shift on hypo-osmoregulatory capacities of smolts regardless of population differences in smoltification timing under conditions without temperature shift. The resilience of such physiological impact was sustained at least for 1 week, comforting the role of high temperature in influencing the rate of changes occurring during smoltification. Therefore, favouring the downstream migration to help smolts reach the sea faster may mitigate the impact of a rapid temperature increase.