Hi Sonya,
These are real effects with the supports arranged as they are as there are local redundancies . For example, in your model you have each support at the second pier fixed in the x direction, which will restrain any longitudinal movement across the deck section at this pier. If the applied load was a simple point load on the mid point of the edge girder in span 2 then you would expect the deck to deform in torsion, transverse bending and longitudinal bending such that the plane section would not remain plane across the whole deck. If the supports are fixed in dx then you will see a reaction at each of these - but the total shouls equal zero if no external horizontal load is applied.
If you reduce your supports so that only rigid body motion is restrained, then you would see this effect in the deflected shape. I suggest that all supports are fixed vertically; support 1202 is fixed dx,dy,dz; and support 1938 is fixed dy,dz. This will provide the minimum supports to resist rigid body motion without introducing any additional local redundancy.
I hope this helps but please let me know how you get on. If this has solved your problem then please mark my reply as a solution so that others may benefit.
Kind regards
Dave Geeves
