If I understand the test fisheries right, they fish at a certain point in the tide every time which is why some days have three if there are three high tides in the day.
It is likely a lot to do with how the river smells. The fish can smell how much fresh rain water is included and fresh rain water would equate to better chance of survival upstream. A number of days of rain will usually see a whole bunch of fish making a mad dash upstream. Days with little rain will see a lot of fish hanging out and waiting while smaller groups will go ahead and chance it.
A river the size of the Fraser, time of day likely doesn't have as much of an effect on when fish enter but it could affect whether they run where the net is or deeper or shallower.
I agree though that it seems counter productive to be removing fish from the system to test how many fish are coming in when there are so few fish. At the same time, it can be an indicator of other problems. If decent numbers are being caught in the test fisheries and then they disappear between there and the spawning grounds, then something is intercepting the fish on the way up. If the river is closed to fishing, maybe not everybody is following that rule. I would think a seine net would make more sense so the fish could be released.