Arrival Time of negative first wave

[matt-mac-ga]

Potential to modify this code to track the arrival time of a positive or negative wave.

ptha/propagation/SWALS/src/shallow_water/domain_mod.f90

Line 2287 in 52abcb2

(domain%U(i,j,STG) > arrival_stage ) .and. &

[gareth-d-ga]

If we do this, then IMO it should be a new optional summary statistic rather than replacing the existing "positive exceedance" arrival time. At the moment the arrival time is not stored by default, the user code has to ask for it, so there could be a few options.

It could be implemented by introducing a new value into gridded_output_variables_max_U (perhaps it could be called "arrival_time_positive_or_negative"). Then we'd have to implement the calculation as a branch of the select case statement near where you've linked above.

One could also consider using a speed or flux threshold -- but it would have to be robust to round-off level noise in the FV solvers, and might be hard to choose a threshold in general.