Capability to calculate the time when the maximum water elevation occurs

[pvelissariou1]

It is desirable in storm surge modeling to know the time when the maximum water elevation occurs (using both internal calculations & post-processing calculations using the ouput files).

[josephzhang8]

Added in: 302907a

Could you please help me test read_output10_allnodes.f90? Thx Takis.

[pvelissariou1]

@josephzhang8 Yes, Joseph. What exactly you want me to do? I am already using read_output10_allnodes in my evaluation scripts. Do I need to update my schism code?

[josephzhang8]

Yes, git pull and recompile that script. No changes in inputs. You should see timing output in *max.gr3 after the maxelev.

[pvelissariou1]

Actually I am using read_output8_allnodes for the shoout* files, read_output10_allnodes for the out2d* files and combine_gr3 to combine the maxelev_[0-9]* files. I'll pull the code and use it in one of my test cases (I believe, I will update you by tomorrow). Thank you for such a quick response.

[josephzhang8]

Thx Takis. I've also added it into read_output8_allnodes in the latest master.

[pvelissariou1]

In both read_output10_allnodes.f90 and read_output8_allnodes.f90 I modified the 'open(12, ... statement to:
open(12,file=trim(adjustl(varname))//'_'//trim(adjustl(outname(k)))//'.gr3',recl=512,status='replace')
otherwise, the times are written in the next line.
As an input to read_output10_allnodes.f90 I am using:

elevation  ! variable name
1          ! switch for variable type: scalar (1) or vector (2)?
1 1        ! start and end stacks; out2d_1.nc, out2d_2.nc, ...
0 1 1      ! switches for min, max, avg; compute stats for 2D var (0/1:min; 0/1:max; 0/1:time avg)
-999999    ! threshold for outputing violators.bp, where violaters are nodes/eles with var values < 
1          ! initialization flag: (0) do nothing;  (1) intialized to -dp (useful for maxelev)

I am checking with 6min outputs and will be plotting the results for TC Helene.

[pvelissariou1]

I'll revert back to formatted output. I am working on a pull request to include combining outputs/maxelev_* and outputs/maxdahv_* files that also include the times at max* values.

[josephzhang8]

In both read_output10_allnodes.f90 and read_output8_allnodes.f90 I modified the 'open(12, ... statement to: open(12,file=trim(adjustl(varname))//'_'//trim(adjustl(outname(k)))//'.gr3',recl=512,status='replace') otherwise, the times are written in the next line. As an input to read_output10_allnodes.f90 I am using:

elevation  ! variable name
1          ! switch for variable type: scalar (1) or vector (2)?
1 1        ! start and end stacks; out2d_1.nc, out2d_2.nc, ...
0 1 1      ! switches for min, max, avg; compute stats for 2D var (0/1:min; 0/1:max; 0/1:time avg)
-999999    ! threshold for outputing violators.bp, where violaters are nodes/eles with var values < 
1          ! initialization flag: (0) do nothing;  (1) intialized to -dp (useful for maxelev)

I am checking with 6min outputs and will be plotting the results for TC Helene.

Why do you need recl=512?

[pvelissariou1]

i had to use the RECL specifier, otherwise the times were written to the next line, instead in the same line with the max. elevation (for unformatted output there is a line length limit of 72 characters I believe. The 512 number is arbitrary. I'll go to formatted output instead. I'll submit a pull request later today with my changes.

[josephzhang8]

Thx. Formatted outputs are the good option.

[pvelissariou1]

Hi Joseph, I created a pull request with all the changes we implemented. Everything looks good from my side. Tested with hurricane helene using, newIO, oldIO and hourly/6min output intervals.

[josephzhang8]

Thx Takis. Please see my reviews.