1. Gross primary production (GPP) and ecosystem respiration (ER) were analy
sed for 18 months in two reaches of the River Thur, a prealpine river in Sw
itzerland. The upper reach at 655 m above sea level (a.s.l.) is bedrock con
strained, has a high slope (0.60%) and a catchment area of 126 km(2). The l
ower reach at 370 m a.s.l. has a more extensive hyporheic zone, a lower slo
pe (0.17%) and a catchment of 1696 km(2).
2. In both reaches, temporal patterns of stream metabolism reflected the oc
currence of bed-moving spates. Average reductions of GPP and ER by spates w
ere 53 and 24% in the upper reach, and 37 and 14% in the lower reach, respe
ctively. The greater resistance of ER than GPP in both reaches shifted the
ecosystem metabolism towards heterotrophy (decrease of the ratio of GPP to
ER (P/R)) following spates.
3. Recovery of GPP was significantly faster in the lower reach and exhibite
d distinct seasonal variation (positive correlation with incident light). T
he differences in stability (both resistance and resilience) between reache
s reflected differences in geomorphic settings and disturbance regime.
4. Stepwise regression analysis was used to explore the potential influence
of season, disturbance and prevailing environmental conditions on stream m
etabolism in each reach. Time since spate plus temperature explained 73 and
86% of variation in ER and GPP, respectively, in the upper reach and 55% o
f variation in ER in the lower reach. Season plus prevailing environmental
conditions explained 67% of variation in GPP in the lower reach.
5. To test how the perception of stability may change with increasing scale
of observation, the disturbance regimes of 12 sites were compared with the
disturbance regime of the entire Thur catchment. The analysis suggests tha
t stream metabolism at the catchment scale is far more resistant to high fl
ow events than at the reach scale.