Episodic Fish Kills in an Acidified Salmon River in Southwestern Norway

Abstract

The objective of this study was to conduct fish kill assessments of Atlantic salmon (Salmo salar) and brown trout (Salmo trutta) and associate the cause of the kill with environmental acidification during the snowmelts. The study was carried out in the river Vikedalselva in southwestern Norway each spring from 1982 to 1985. River acidity was measured daily in 1982 and continually from 1983 to 1985. Fish kill was assessed daily by inspecting stretches of the river bottom. A total of 680 carcasses of Atlantic salmon and brown trout were either recovered or counted. Most of the dead specimens were juvenile fish, of which a predominant fraction were ≥ 2 years old (presmolts). Mortality varied considerably between years, from a high of 376 specimens in 1984 to a low of 34 in 1985. Another 39 specimens were observed drifting in the river. The fish kills were generally episodic, and usually followed sudden increases in discharge and depressions of river pH. With the exception of 1983, a correlation analysis using linear regression revealed a significant inverse relationship between daily pH and the number of dead salmon and trout recovered. Depression of pH did not cause instantaneous death, and peak abundance of carcasses found was delayed relative to the initiation of these episodes. This is probably because fish became stressed prior to death, as indicated by observations of downstream drifting of bodies. Fish kill seems to be related to prevailing runoff conditions, i.e., in 1985, when runoff was low due to thin snow cover and lesser amounts of rain, mortality was insignificant. This study indicates a correlation between pH and mortality of Atlantic salmon and brown trout. However, such field studies are also subjected to some irregularities between river acidity and fish kill. This may be related to the method used for assessing the fish kill, or to hydrological as well as biological factors.