Added some small changes to the FTLE code, as described in email to Knut-Frode. The fields look similar, but values now tend to be more positive. This is more apparent for model data than for the double gyre example.
Not sure if values are 100% correct now (or if they even matter).
I expect that majority of values in a domain of real flow should be positive, as negative FTLE values describe non-chaotic flow.
Added some small changes to the FTLE code, as described in email to Knut-Frode. The fields look similar, but values now tend to be more positive. This is more apparent for model data than for the double gyre example. Not sure if values are 100% correct now (or if they even matter). I expect that majority of values in a domain of real flow should be positive, as negative FTLE values describe non-chaotic flow.