On a technique to investigate the covariation between streamflow and rainfall in afforested catchments

Principal components analysis is used to split the variance of streamflow and rainfall in three catchments into uncorrelated factors in an attempt to describe the effect of afforestation on streamflow. The total period covered is from 1956 to 1977, and that up to 1969 was treated as a control period. Most of the variation in streamflow is associated directly with rainfall for Pinus patula and the control area; this is not, however, true of the Eucalyptus grandis catchment which has an anomalous time series which has undergone a steady decline over the whole period of record. It is suggested that there is a principal component that isolates the effects of having planted the P. patula in 1971.