… multiple parameters user changeable (snapwave_gammaig, snapwave_Tinc2ig, snapwave_shinc2ig,snapwave_finc2ig) as in HurryWave. Added factor to control offshore estimate of IG energy: snapwave_eeinc2ig

- Change default depth of snapwave from 0.01 to 0.001m for more flexibility and accuracy in specifying snapwave_hmin

- Make possible to change the baldock_ratio (default = 0.2 as before)

…'dw'and 'df' if storefw = 1

…waves to his-output as 'dw'and 'df' if storefw = 1

- TODO; still check better whether integrated calculation is correct

- Add Battjes-Janssen 1978 breaking dissipation model for IG waves with 2 options to apply breaking everywhere or where H>gamma*depth, controlled by snapwave_battjesjanssen_opt=1 (default) or 2
- add snapwave_ig_opt to swith between Baldock for IG waves =1 (default) of BattjesJanssen

…previous timestep as in Baldocks formulation and write to sfincs netcdf output

- Calculate directionally averaged normalised bed slope betan and write to sfincs netcdf output
- Calculate directionally averaged shoaling factor fsh and write to sfincs netcdf output
- Make fsh be dependent on one minus the percentage of breaking waves Qb instead of assuming Qb=1 as in Hurrywave implementation
- Make coefficients in fsh user changeable and have a minimum value as in Hurrywave implementation -> snapwave_fshfac=1, snapwave_fshexp=15, snapwave_fshalphamin=0.1
- Add a coefficient that controls how much of the incident energy sink ends up as IG wave source -> snapwave_shpercig=0.25 -> meaning that not all dissipated energy goes to the IG frequencies
- Bugfix so Hmx_ig is actually controlled by gamma_ig instead of gamma of incident waves
- Some intermediate attempts at different options inside Baldock breaker formulation

… waves

- add user parameter snapwave_depthforcerelease controlling below what water depth, there will be no inc2ig energy transfer anymore

…gy source is given anymore to IG waves, should become more generalised later

…_snapwave_ig

# Conflicts:
#	source/src/sfincs_lib.f90
#	source/third_party_open/netcdf/netcdf4.vcxproj

… monochromatic

- Propose tp = snapwave_tpmean*snapwave_Tinc2ig instead of fixed value of 6
- Was needed to read in snapwave_Tinc2ig again because snapwave_data is not available in SFINCS itself

- Add 'srcsh' variable of energy sink incident waves/energy source ig waves to his output, is directionally averaged
- Therefore, in snapwave/snapwave_solver.f90 change srcsh into srcsh_local

…o between Tp,inc and Tp,ig: snapwave_Tinc2ig = 7 (default) instead of hardcoded ratio of 6.0

- add quadtree print statements as in main

- remove unused 'igperiod' variable

…sh formula in subroutine 'estimate_shoaling_rate', still with regular Baldock wave dissipation

- Option of old fsh formula with BattjesJanssen IG dissiptation is ig_opt = 3

- Reset defaults of shinc2ig and shpercig

- Still have problem at boundaries
- Still need to calcullate steepnes, reldepth, depthforcerelease based on weighted nearest offshore wave data

- Use now downwind IG wave height from previous timestep, to evaluate choice still

- if (Sxxprev(itheta)<=0.0) then is enough

- Note the srcsh / Cg(i)

…d values of 8

- Not sure whether all real*8 were needed in the end...

…shold helps

- Still to be reviewed whether method works good like this with all in rms instead of Hm0

… to limit 'inverse' shadow zone for IG waves, induced by the srcsh component

- The -1 is now accessible out of ierr = sfincs_update loop when running through BMI I think
- However, not yet as windows exe error message, since still the program stops normally, requires more work
- Note, branch is split from current main, which had some small issues in quadtree netcdf when compiling in debug mode

…already helps a lot

…better0, and then don't have to limit dSxx to min 0

…ent wave heights for calculating the offshore wave steepness, representative waterdepth, and alphaig shoaling parameter depth cutoff

- Directly divided by sqrt(2) to go from Hm0 to Hrms
- In initial test with multiple boundary input points with different input wave heights implementation seems to work

…ad of being a user input

- Better to not divide by sqrt(2) yet in snapwave_boundaries.f90>update_boundaries

…th Cg/h

- rename into shoaling_parameter

… alphaigfac factor in case needed

- Fix big bug in alhaig parameter fitting (xbeta and ybeta needed)
- Output betar instead of betan

…t to 10*Hm0,inc

- This makes sure that e.g. no transfer is initiated in 200m water depth, with constant alphaig=8
- Add timer of SnapWave energy balance subroutine again

…/cg!

- Do deepwater cutoff now at reldepth=20
- Add min value for H_rep tomake sure that H_rep(k) is never 0! (problem in calculating steepness_bc, reldepth and depthforcerelease later)

… wave direction itheta

- srch should not be /Cgi+1 but /ds

…y needed

- Do check that reldepth is > 0

…g by factor 2

…ctor in the Baldock wave dissipation formula, so its separate from incident waves

- Add a separate user option for baldock_ratio_ig, so its separate from incident waves
- Clean up code a bit by removing ig_opt 2,3,4 and old parametrisations
- Tried real*8 for beta, betan_local, betar_local, not sure yet it is needed/helps
- For now calculate betan output as maxval, not sum like alphaig

…ee_ig, because ee is already Hrms based

- Also fix change b to beta
- Fix gamma using Hinc_old from previous timestep

- Massive but weird bug - for some reason (k1*k2)>0 is not reliable always, resulting in directions being uncorrectly skipped!!! solved with:   if (k1>0 .and. k2>0) then ! IMPORTANT

…ent waves. parameter: snapwave_shinc2ig = 1 to subtract. Use snapwave_shinc2ig = 0 (default) to only add energy to the IG waves, without incident wave sink

…anges with wave setup/surge/tide, and therefore als beta, which is not what we want I think

…ing conservative shoaling, with snapwave_ig_opt = 2. For the rest everything is the same as with _opt = 1

…(and a Tpig), for boundary points and timesteps is added, this is all processed correctly in subroutines 'update_boundary_points', 'update_boundaries' and as input to snapwave_solver.f90

- Now ee_ig is build outside of solve_energy_balance2Dstat already, and can be spatially varying
- Now T_ig is determined outside of solve_energy_balance2Dstat already, grid mean value

… latest local water depth

…ind_nearest_depth_for_boundary_points() > needed for Herbers estimate of Hig0

- Code seems ready to add actual routine into 'build_boundw', next year!

…ng correctly

- Working now is making jonswap spectrum in build_jonswap

…swap

… of build_jonswap subroutine of XBeach' waveparams.F90

- think something with 'theta' and directions is too complicated now and not working
- And something funny with those 'double allocated vars'

…nswap

- End of Part 2, loosely based on XBeach subroutine 'build_etdir' from waveparams.F90
- Used 2D interp insteadof XBeach code

…o input Hm0, as in XBeach option correctHm0 =1)

- Default is off (=0) as in Matlab&XBeach code

…e_boundary_conditions

- Important! Overruled therefore this line in snapwave_boundaries.f90 - might conflict with running SnapWave standalone
- Catch unrealistic low/high calculated values for Hm0ig

- Added possibility to make choise what period to choose for Tp later user changeable

… code

…ctions in a separate module and file!

- Also added to Makefile

…eterisation versions

- Added option whether one wants to use the herbers determined Hm0ig&Tp for boundary conditions (snapwave_use_herbers = 1, default)
- Added option then for choosing Tpig type method based on computed spectrum (snapwave_tpig_opt = 1, default). Options are: 1=Tm01 (default), 2=Tpsmooth, 3=Tp, 4=Tm-1,0

Other notes - SFINCS changes:
Changed: references of betan are now beta in SFINCS netcdf output
Removed: option that SFINCS wavemaker directly uses the snapwave_tpmean * snapwave_Tinc2ig > now let it come from Snapwave,
where snapwave_tpigmean comes from (still there the option to user define snapwave_Tinc2ig, using snapwave_tpigopt=XX)

Other notes - SnapWave changes:
Removed: option in find_boundary_indices() that would find the nearest grid active point (kcs=1), that was used before in case of IG waves
Removed: old vars like H_rep, steepnes_bc, depthforcerelease etc.
Removed now unused variables: snapwave_shpercig, snapwave_fshalphamin, snapwave_fshfac, snapwave_fshexp
Kept: baldock_opt > discuss later whether we want to keep this with SnapWave team

- Add all snapwave vars also to attribute list for netcdf output, not all possible because of same name as in sfincs_data. Can see whether/how to add those later
- renamed iig to igwaves_opt, and put in snapwave_data
- Removed write statement, tpinc and tpig values are available in sfincs_his.nc now by default

…v_ig' in sfincs_snapwave.f90 after call update_boundary_conditions & compute_wave_field so values at first timestep are not 0!

- Also added check on incoming Tpig in sfincs_snapwave, that is used in SFINCS' wavemaker routine.

- Add check and warning on water depth in snapwave_infragravity - determine_ig_bc, because if you have snapwave.bnd points in shallow water (e.g. <10m), and expecially with low directional spreading, the computed Hm0ig can be large (>2m)
- Consider calculating a mean deptht_bwv to use for herbers if deptht_bwv(ib) is very low, or set back to 10 m minimum. For now it's only a warning

…_snapwave_ig

-  zsm also used in sfincs_continuity if 'snapwave=true' > todo: check if needed, or only for wavemaker
- remove reduction of IG wave height at boundary in case of high values in Herbers - for now

fixed issue where combined uv indices were not set properly (issue only occurred icm snapwave)
added advection_mask (set to 0 next to boundaries and wave makers)
added check on time series for wave makers and snap wave not covering entire simulation time span

…...)

- fix some typos