Geomorphic comparison of two Atlantic coastal rivers
Substrate size and mobility are important to Atlantic salmon (Salmo salar) spawning and rearing success. Channel geometry is a control on bedload mobility in streams. It is believed that channel morphology in many Maine rivers has been altered by land use practices, creating wider and shallower channels, and lowering stream competence. If correct, these changes may be partially responsible for the limited number of returning salmon currently observed in Maine coastal rivers. To evaluate the magnitude of these changes, I performed a statistical comparison of channel morphology between two Atlantic coastal streams: the Narraguagus River in Downeast Maine and the Jacquet River in northern New Brunswick, Canada. Compared to the Narraguagus River, the Jacquet River has relatively healthy returns of adult salmon. Both watersheds have similar drainage areas (Narraguagus 588 km²; Jacquet 510 km²;) and mean annual precipitation (1244 mm; 1200 mm), but differing average channel gradients (0.16%; 0.51%) and longitudinal profiles. During the summer of 2007, I surveyed a 13.6-km section of the Narraguagus with a drainage area range of 129-247 km², and a 10.4-km section of the Jacquet with a drainage area range of 94-265 km². I made measurements of active and bankfull width and depth, and channel gradient at 100-m intervals, and performed grain-size counts at 200-m intervals. I also measured gradient and width in a GIS-based analysis. Results of my analysis show that channel gradient is likely the most influential factor on Atlantic salmon habitat as it relates to sediment size. The two rivers exhibit no significant difference in width-to-depth ratio, when low-gradient outliers in the Narraguagus River are removed. Predicted median riverbed grain sizes were calculated using two methods: (1) from the empirical relationship between basal shear stress and measured grain size; and (2) using the Shields parameter and remote sensing data only. Measured and predicted grain sizes reveal finer river-bed sediments on the Narraguagus River, however, Shields parameter calculations show that sediment should be mobile in both streams. I compare these predictions to field-based habitat mapping on the Narraguagus River. Based on predicted grain sizes, I expect nearly continuous Atlantic salmon spawning (28-95%) and rearing (95-100%) habitat on the Jacquet, and much less spawning (47-62%) and rearing (57-68%) habitat on the Narraguagus. This is likely because the Narraguagus River is segmented into reaches of steeper gradient (S < 0.002) with potentially good habitat, and flatter reaches (S < 0.0005) of poor habitat. The long flat reaches (several km) likely act as sediment sinks, preventing the continuity of downstream sediment transport and causing sediment to be sourced from localized glacial deposits.