Lagrangian floats

[onken]

In ROMS, I have launched various isobaric floats at Fz0=-1.00 m depth. In the floats output file, however, the variable depth is around -0.85 m all the time:

depth =
_, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _,
_, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _,
_, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _, _,
-0.8510733, -0.8510394, -0.8510075, -0.8509726, -0.8509416, -0.8509079,
-0.8508808, -0.850847, -0.8508143, -0.8507733, -0.850736, -0.8506947,
-0.8506593, -0.8510703, -0.8510303, -0.8509935, -0.8509398, -0.850899,
-0.8508561, -0.8508234, -0.8507994, -0.8507763, -0.8507558, -0.8507443,
-0.8507453, -0.8507468, -0.8511912, -0.8511894, -0.8511964, -0.851199,
-0.8511897, -0.8511713, -0.8511428, -0.8510867, -0.8510244, -0.8509563,
-0.8508586, -0.8507845, -0.8506688, -0.850555, -0.8508522, -0.850715,
-0.8505963, -0.8504642, -0.8503275, -0.8502064, -0.8500525, -0.8499502,
-0.8498107, -0.8497502, -0.8496121, -0.8495278, -0.8493841, -0.8496665,
-0.8495473, -0.8494577, -0.8493702, -0.8492821, -0.8492153, -0.8491327,
-0.8490607, -0.8489813, -0.8489082, -0.848835, -0.8487664, -0.8487049,
-0.848996,
-0.846485, -0.8465011, -0.8465173, -0.8465326, -0.8465464, -0.8465579,
-0.8465666, -0.846571, -0.8465714, -0.8465693, -0.8465682, -0.8465698,
.... .... ....
...


Any idea what's wrong? Please find attached the floats input file float.in and the log file log.txt.
By the way, when I make the floats Lagrangian (T=1), the initial depth is again at -0.85 m.
Thanks, Reiner

[wilkin]

What's the typical value of zeta (sea level). Might be something to do with depth being with respect to z=0.

Did you verify the depth was correct at initialization and then changed? Or is it starting out wrong.

[onken]

Oh yes - you are right! I initialized ROMS from a larger domain by downscaling, and zeta is around 0.15 m. That's exactly the missing difference!
Thanks, Reiner