Observation dates seem off

[Huite]

See image of this drawdown. t=0 results in the right value (10.0), but it should be placed at Monday 15 0:00, not at 2:00.

All the dates seem somewhat shifted.

[CGoudswaard]

Another quirk of the observation dates appears to be a discrepancy between how calculation points are placed in time. In the following situation, the filters start at t = 1, the Head Observation is set at t = 0,9 and t = 1,9, and the ttim Computations times are set at t = 1 and t = 2.

What appears to happen is Head Observations are attributed to the time period up to t = 1, whereas the raster calculation is attributed to to the time period following the computation point ( t = 1 to t = 2 ). This results in the contour plot 'lagging' the head monitor by 1 day:

It would appear logical to me if the Head Observation points are attributed to the time period following the calculation moment (like the rasters), not the preceding period.