Canada ; Cold area ; Comparative study ; Core sampling ; Freezing ; Gaspé ; Quebec ; Research technique ; River bed ; Sampling ; Statistical bias
The aims of this paper are to quantify the relative biases of the freeze-core and bulk sampling techniques against more accurate composite samples and to examine whether systematic losses of fines can be predicted. At 6 riffles the spawning
substrate was sampled under approximately 50 cm of water with a bulk sample and 3 adjacent freeze-cores. For each riffle, data from the 2 samples were combined using the method of Fripp and Diplas (1993) and the resultant composite sample was compared
with the original freeze-core and bulk samples to assess the relative precision and biases of the 2 techniques.