I have another question about the shear training data at 30m and 10m. I also saw that the 10m mean value is significantly lower than the 30m mean value for sets 2 & 3, which is expected because of the low height and significant wind shear. However for shear training sets 1, 4 & 5 the 10m mean is very similar to the 30m mean. In fact, for sets 4 & 5 the 10m mean is greater than the 30m mean a significant portion of the time. This seems rather strange. Can anyone confirm this observation and if so would anyone like to comment?