DEVELOPMENT AND APPLICATION OF A NUTRIENT-DIFFUSING BIOASSAY FOR LARGE RIVERS

1. Laboratory and field experiments were performed to develop and then apply a nutrient-diffusing substratum (NDS) design suitable for use i n large, fast-flowing rivers. 2. Initial laboratory experiments quanti fied diffusion of PO4 and NO3 from new and previously used clay pots, which were soaked in deionized distilled water. Mean release rates ini tially exceeded 2.4 and 725 mu mol l(-1) day(-1) P and 0.22 and 18 mu mol l(-1) day(-1) N from new and used pots, respectively, but declined rapidly with increasing time spent in deionized distilled water and w ere below detectable levels after about 18 and 29 days, respectively. 3. A phosphorus (P) dose-response experiment in a P-limited reach of t he Athabasca River, Alberta, Canada showed that epilithic biomass and macroinvertebrate density on NDS increased with increasing concentrati ons of KH2PO4 up to about 0.5 M. Beyond this threshold, biomasses and densities were unaffected by initial KH2PO4 concentration. Coefficient s of variation of epilithic biomass estimates declined with increasing KH2PO4 whereas invertebrate density appeared to be unaffected by KH2P O4 levels. 4. Release rates of both P and N from NDS filled with 0.5 M KH2PO4 or 0.5 M NaNO3 declined at a log-negative rate from about 5000 mu mol N-NO3 l(-1) day(-1) and 3500 mu mol P-PO(4)l(-1) day(-1) on da y 2, to 200 mu mol l(-1) day(-1) for both N and P on day 32. 5. After development, we used the diffusing substrata to identify spatial patte rns in nutrient limitation at seven sites along a 120 km reach in the Athabasca River, that receives two known point-source nutrient inputs. NDS consisting of N, P, N + P and unenriched controls were attached t o the river bottom for 22-23 days and then retrieved and sampled for e pilithic chlorophyll a. Physicochemical parameters and epilithic bioma sses on upper stone surfaces were also quantified when NDS were deploy ed and retrieved from each site. 6. Sites located immediately downstre am of the two point source inputs had higher water column concentratio ns of PO4 and epilithic biomasses than the site immediately upstream; epilithic biomass was positively related to PO4 in the late autumn (r( 2) = 0.58) but not in early autumn. Sites located immediately below nu trient inputs were not nutrient-limited, whereas upstream reference si tes were P-limited.