A stable and accurate data assmimilation technique using multiple penalty terms in space and time

• A new method for data assimilation based on weak imposition of external data is introduced. The technique is simple, easy to implement, and the resulting numerical scheme is unconditionally stable. Numerical experiments show that the error growth naturally present in long term simulations can be prevented by using the new technique.

Ämnesord

NATURVETENSKAP  -- Matematik -- Beräkningsmatematik (hsv//swe)

NATURAL SCIENCES  -- Mathematics -- Computational Mathematics (hsv//eng)

Nyckelord

data assimilation

summation-by-parts

weak boundary conditions

multiple penalties

stability

finite differences

Publikations- och innehållstyp

vet (ämneskategori)

rap (ämneskategori)