It is often necessary to estimate extreme events at sites where little or n
o hydrometric data are available. In such cases, one may use a regional est
imation procedure, utilizing data available from other stations in the same
hydrologic region. In general, a regional flood frequency procedure consis
ts of two steps, delineation of hydrologically homogeneous regions and regi
onal estimation. This paper focuses on the development of a regional flood
frequency procedure based on canonical correlation analysis (CCA) and its a
pplication to data from a northern Canadian basin in which hoods are domina
ted by spring snowmelt. This CCA-based procedure allows the joint regional
estimation of spring hood peaks and volumes. The CCA method allows the dete
rmination of pairs of canonical variables such that the correlation between
the canonical variables of one pair is maximized and between the variables
of different pairs is equal to zero. Therefore, it is possible to infer hy
drological canonical variables, knowing the physiographical-meteorological
canonical variables. The methodology developed was applied to the St. Mauri
ce River basin system, which is operated by Hydro-Quebec and characterized
by the relatively low precision of flow data available. Results show that t
he proposed method allows for a significant reduction in the 100-year sprin
g flood and volume quantile estimation bias and mean square error. The stud
y also shows that, in 60% of cases, the method that was previously used ove
restimates quantile values, which leads to an overdesign of retention struc
tures.