S. Lamontagne et Sl. Schiff, The response of a heterogeneous upland boreal shield catchment to a short term NO3- addition, ECOSYSTEMS, 2(5), 1999, pp. 460-473
Boreal Shield rocky ridges at the Experimental Lakes Area, northwestern Ont
ario, contain two plant/soil communities with contrasting N cycles. Picea m
ariana-Pinus banksiana "forest islands" are N limited whereas the lichen, m
oss, and grass community (or "lichen patches") on the surrounding bedrock o
utcrops appear intrinsically N saturated. The potential for this landscape
to retain a N input of eightfold ambient levels was tested with a 2-y addit
ion of 40 kpa N ha(-1) y(-1) as NaNO3 to one small catchment (0.40 ha). The
elevated N input was poorly retained by the whole catchment during snowmel
t. However, during the growing season, N retention in the treated catchment
remained as efficient as in references. Forest islands and bedrock surface
s responded in opposite fashions to the elevated N input. By the second yea
r of N addition, bedrock surfaces no longer retained additional N inputs. I
n contrast, N-amended and reference forest islands retained a similar propo
rtion of N inputs, indicating that forest islands did not become N saturate
d. The response of the whole catchment to N addition was more similar to fo
rest islands than bedrock surfaces. Even if forest islands only cover a sma
ll proportion of catchment area, they can have a strong impact on whole cat
chment element export because most of the water must move through at least
one island before leaving the system. Because the different components of t
he boreal shield landscape are hydrologically connected, N saturation may o
ccur as a cascading effect in this ecosystem. Monitoring boreal shield land
scapes by using outlets at the lower end of the hydrological cascade can fa
il to detect the impacts of perturbations such as increased N deposition on
upper components.