logsheet.txt

== Version 5.3.11 (08/19/2024) ==
* Changed to alternate between recurring hot start files.
* Added folder to start holding test cases for Continuous Integration

== Version 5.3.10 (07/16/2024) ==
* Added compile option for VS2022, Intel OneAPI 2024.1
* Fix for an erroneous Save Point warning that was always written.
* Improved status description update for Culverts in CMS.
* Added ERROR check for missing X/Y Origin cards in .cmcards file.
* Implemented an extra file to store Save Point cards when there are more than 10 defined.

== Version 5.3.9 (04/16/2024) ==
* Fix for Dredge module diagnostic file not found.
* Clarified some diagnostic output regarding simulation duration and computed residuals.
* Fix for steering variable (NHV_str to NH_str) possibly related to Flow/Wave steering issue.

== Version 5.3.8 (01/08/2024) == 
* Update for internal path length for tidal database file locations.
* Investigating an issue with internal mapping from UTM to Lat/Long in tidal database cases. 
** Use State Plane if possible.

== Version 5.3.7.1 (12/20/2023) ==
* Began adding CF Compliant naming to HDF5 solution datasets (not yet implemented).
* Incremented CMS-Wave version number to 3.3. The code was modified in 2021 but the number was not incremented.
* Fix for Tidal boundaries where Offset used - it was doubling the offset.
* Getting ready for Open Source - several files are modified/moved.
* BugFix in Steering for ADCIRC tidal databases with waves.
* BugFix for some tidal database forcing issues.

== Version 5.3.6 (09/26/2023) ==
* Minor fixes and diagnostic output improvements for Rubble Mound Jetties with SMS 13.3+.
* Incorporate fixes for some explicit issues

== Version 5.3.5 (09/08/2023) ==
* Fix issues with Tidal Database operation with CMS. Improved screen output.
* Fix issue with Grid Angle introduced with SMS 13.1.
* Minor GFortran fixes for compiling on HPC/Linux with newer Gnu compiler.

== Version 5.3.4 (05/16/2023) == 
* Bug fix for missing zero output time in files when increment > 100 hours.
* Added a warning if user-specified boundary angle was too different than internally calculated angle for that boundary.
* Fix in CMS Wave GSR Solver to avoid an infinity in certain situations.

== Version 5.3.3 (04/07/2023) ==
* Split third-party code (spatial transformations) out as separate linkable libraries needed to Open Source CMS (in process).
* Implemented two new ADCIRC tidal databases (EC2015 and ENPAC2015).  Currently in testing.

== Version 5.3.2 (08/11/2022) == 
* Minor change to reading parameter file for Explicit scheme.
* Bug fix when choosing multi-sediment with D35,D50,D90. Fixed array allocation error.
* Bug fix in one lookup-table, 'bs_init' routine.
* Bug fix in Tools (Option 4) when merging datasets written by SMS instead of CMS solution datasets.

== Version 5.3.1 (07/07/2022) ==
* Minor change to Weir Structure specification cards for integration into SMS 13.x (13.2 at present).
** Add option to specify Cell IDs in the same manner as for Rubble Mound Jetties.
* Updated the array upper limit for several CMS-Wave variables.

== Version 5.3.0 (05/16/2022) == 
* Initial version of CMS to be released with SMS version 13.2
* Change cardnames and documentation from 'CSHORE' to 'C2SHORE'.  Upcoming SMS 13.2 interface will reference 'C2SHORE_xxx' cards.
** C2SHORE is a new option for Cross-shore sediment transport.  It requires the CMS to be run with both waves and flow. CMS will fail if both waves and flow are not specified.

== Version 5.2.18 (04/29/2022) ==
* Bug fix for DREDGE_RATE card in the dredging module. Was not reading units as it should according to documentation.
* Bug fix for divide by zero in WVBRK_Inline - rewrote using conditionals instead of GOTOs and also correcting the possibility of a divide by zero.
* Bug fix for using Initial Condition file outside the current working directory.
* New card 'EXTEND_DURATION_RUN_FOR_HOTSTART' with value 'ON'|'OFF' has been added. This will extend the selected DURATION_RUN by the time written in the IC File.
* Minor - force user to associate datasets with their choice of Friction type (Manning's, BF, or Roughness Height)
* Minor - added better diagnostic output for Dredge Rate showing both daily and per second rates.

== Version 5.2.17 (03/29/2022) ==
* Bug fix for Wave structures
* Bug fix for Dredge/Placement where no cards exist between PLACEMENT_BEGIN and PLACEMENT_END statements.

== Version 5.2.16 (02/23/2022) == 
* Bug fix - Path was not being prepended to the wave file names, so if they weren't in the same directory, it failed.
* Added new option to 'Tools' menu for merging datasets from multiple files into one solution file (XMDF).

== Version 5.2.15 (01/21/2022) ==
* Bug fix - Error while using Thickness datasets for bed layers 
* Bug fix - Variables not allocated when Tecplot output enabled.
* Bug fix - Divide by zero in some situations during Bottom Friction routines.
* Bug fix - Wave Date output in screen/diagnostic output. Changed to 'YYYY-MM-DD HH:MM:ss UTC' format
* Minor - added better diagnostic output for Avalanching repose angles too large.
* Minor - added better diagnostic output when error on reading Wave simulation file.

== Version 5.2.14 (01/07/2022) ==
* Minor: Added extra line of output to screen and diagnostic output while Ramp is in effect with the percentage being applied.
* Minor: Added extra code to ensure the acceptance of various versions of horizontal projection
** i.e., NAD83, NAD_1983, NAD1983, NAD_1983_PA11, and so forth.
* Bug fix: Added a few extra initialization steps to Wave code
* Bug fix: Divide by zero when calculating wall friction coefficient.

== Version 5.2.13 (12/06/2021) ==
* Minor: Modification to new CMS-Wave .std format for a few cards
** New card is 'WV_ROLLER_EFFECT' and it takes the options: 'OFF', '25_PERCENT','50_PERCENT','75_PERCENT', and '100_PERCENT'
** New card is 'WV_LIMIT_OBSERVATION_OUTPUT' to turn off optional output files when observation cells are specified.
** For card 'WV_ENABLE_WIND', a new option exists - 'ON-LIMIT_WAVE_INFLATION' 
** For card 'WV_BREAKING_FORMULA', two new options exist - 'MICHE_ORIGINAL' and 'LIFTING_BREAKING'
** Added capability to handle 'HORIZONTAL_PROJECTION' and 'VERTICAL_PROJECTION' blocks (skip over all lines)
** Change options for WV_BOUNDARY_NESTING to "AVERAGE_SPECTRA" and "INVERSE_DISTANCE" instead of "LINEAR" and "MORPHIC"
* Bug fix: Small fix for WV_NESTING_CELLS and WV_OBSERVATION_CELLS

== Version 5.2.12 (11/16/2021) == 
* Minor: Repo fix for VS2013 GUI to make consistent
* Minor: Updated 'CMakeLists.txt' file for Linux
* Bug fix: Corrected an error reading in the WV_NESTING_CELLS and WV_OBSERVATION_CELLS cards.
* Bug fix: Wave code was still looking for a friction file when a constant was being specified.
* Bug fix: Dimensioning of variables when backward reflection enables was sometimes not correct.

== Version 5.2.11 (11/08/2021) ==
* Bug fix: Corrected an error reading in the WV_NESTING_CELLS

== Version 5.2.10 (10/26/2021) ==
* Added new CMS-Wave input format for options file (.std) similar to how the Flow .cmcards file is written. 
* Added a new variable for user-definable gamma for the BATTJES-JANSSEN 1978 breaking formulation.
** WV_SET_GAMMA_BJ78 is the new card to specify that value and works only for BATTJES-JANSSEN 1978.
* DLL for this version and forward must be a newer one that supports the Intel OneAPI OpenMP library.
* Card IGNORE_BRK_RESTRICTION no longer needed to use any Wave Breaking criteria.

== Version 5.2.9 (09/30/2021) == 
* Max WSE Tool now works for ASCII file as well as XMDF

== Version 5.2.8 (09/29/2021) ==
* Added new option to 'Tools' menu for post-processing the CMS Water elevation file to create a Maximum WSE file/dataset.

== Version 5.2.7 (09/20/2021) ==
* Minor: Added more diagnostic output for various types of Derivative Calculation routines.
* Added output option to tell CMS to output maximum WSE values over simulation period for each cell.
** New card name for this option is "MAXIMUM_WATER_LEVEL_OUTPUT" and takes 'ON' or 'OFF' as arguments.

== Version 5.2.6 (06/25/2021) == 
* Bug fix: Fixed error if FALL_VELOCITY, CRITICAL_SHEAR, and COREY_SHAPE_FACTOR cards had no value.
* Minor: Added descriptive print statements to structure types instead of an application error.
* Minor: Added error message when the Wind file format is not an expected type.
* Bug fix: Small fix unset the TELFILE name when a GRID_FILE of '.cart.' type is found.
* Added function to parse through a cardfile looking for a specific card and returning the rest of the line.

== Version 5.2.5 (06/10/2021) == 
* Bug fix: Hot start with multiple sediment grain sizes was broken due to bad names in path to dataset.
* Bug fix: Using ASCII input files wasn't working for salinity. 
* Bug fix: Small logic error in boundary condition corrected.
* Bug fix: Corrected sign error which resulted in a sediment bed-change inconsistency 
* Bug fix: Initialization was missing for initial condition with ASCII input file for Vector data types.
* Minor: Added capability to write out both XMDF and ASCII output files by using "OUTPUT_FILE_TYPE    BOTH"
* Minor: Ensured both cards for running CMS with initial conditions were excluded from a warning.
* Minor: Removed fix for ILUTP solver which caused other issues.
* Minor: Clearer error messages in reading from input files.
* Added: Command line option 'Tools' to perform a few independent tasks.
** Added two routines - one to convert an actual date to a REFTIME and one to do the reverse.

== Version 5.2.4 (06/10/2021) ==
* Bug fix: When using single wave condition, CMS now stores the gridname in a new variable for proper internal reference.
* Bug fix: CMS now stores ADCIRC Parent grid and path information correctly for nesting.
* Bug fix: Divide by zero during normalization of river flux boundaries corrected.
* Bug fix: CMS was looking for a ZONE when Geographic coordinate system was specified.
* Bug fix: CMS only recognized NAD83/NAD27/LOCAL for Datum. Modified to accept any of the other variants (NAD 1983, etc)
* Bug fix: ADCIRC fort.64 file now uses -9.9999e+04 for dry cells instead of 0.0 m/s. CMS needed a small fix to read properly.
* Bug fix: STARTING_JDATE uses a 2-digit year. Fixed logic to make years > 50 = 1900+yr and all other years 2000+yr. (change of 20 to 50)

== Version 5.2.3 (09/10/2020) ==
* Minor: added optional argument to 'time_sec2str' for outputting seconds with I0 or F7.3

== Version 5.2.2.0 (09/04/2020) ==
* Incorporated some bug fixes and needed this number for coordingating with Aquaveo on SMS release.
* Bug fix: Isolated ocean cell checking had an issue. Removed extra output file, switched to screen output.

== Version 5.2.0 (07/31/2020) ==
* Initial release of 5.2 compatible with SMS version 13.1.

== Version 5.1.14 (02/26/2020)==
* Bug fix: Avoid accessing unallocated array in some situations (boundaries.F90)
* Change: Increased number of reflection cells allowed and consolidated this value between inline and stand-alone.
* Change: Improved handling of Isolated ocean cells - output all values to a text file instead of showing only the first, then STOP.

== Version 5.1.13 (01/07/2020)==
* Bug fix: CMCARDS was only able to read a maximum of 1000 lines.  Anything else was ignored.
* Change: Improved printing of information regarding Statistics process types.
* New: Added the ability to put advanced cards into 'advanced.cmcards' for processing if desired.

== Version 5.1.12 (10/25/2019)==
* Bug fix: Corrected display of Rubble Mound Structure cells in diagnostic output
* Bug fix: Removed extra output file of Hard Bottom warning cells
* Change: Started adding in extra code for NetCDF input/output files (logical operators, etc).
* Change: Increased the limit to the maximum number of save points from 255 to 1000.  Added error messages.

== Version 5.1.11 (09/06/2019)==
* Change: Moved all repository folders up one level, removing "Trunk" folder which was not needed.
* Change: If any Unknown Cards are read from the parameter file, CMS gives the user a chance to stop and fix.
* Change: Added __linux definition to CMake options.
* Change: Output message when Boundary Cellstring is defined, but lacks cells (SMS issue)
* Bug fix: Corrected infinite loop experienced in Linux during interpretation of command line arguments.
* Bug fix: Fixed issue with d16/d50/d84 percentile allocatable array
* Bug fix: Fixed issue with missing Hard Bottom Dataset
* New Feature: Dredge Module is now available using ASCII input/output.

== Version 5.1.10 (08/29/2019)== 
* Change: Modified screen and diagnostic file output by aligning at T40. Mainly for Implicit runs.
* Change: Added years to reoccuring report of elapsed time.  Only outputs when Days > 365.
* Change: Updated instructions for CMake linux utility regarding include file changes.

== Version 5.1.9 (08/27/2019)==
* Bug fix: Heatflux Windows vs Linux options and output.  
* Bug fix: Corrected issue with `atan2` operation in inline/standalone wave code.
* Bug fix: Issue with reading in OWI formatted .win and .pre files.
* Change: Removed the requirement for the .xy file to be used with OWI wind format - still allowed if present.
* Change: Updated CSHORE implementation with user-specified parameters.
* Change: Modifications needed for Dredge Module and interface.  
* Change: Removed compile-time option for the Dredge Module.  
* Change: Eliminated need to change compile-time options depending on Linux or Windows machines.
* Info: A few other minor modifications.

== Version 5.1.8 (03/12/2019)==
* Bug fix: Needed to check that a boundary array was already allocated before an allocation in rare cases.
* Bug fix: Implemented a fix for a rare 'divide by zero' error.
* Bug fix: Hot Start divide by zero error corrected.
* Bug fix: Issue with simplified multiple sediment transport definition thinking bed layer was already defined.
* New feature: Implemented variable Morphology Acceleration Factor using a user-defined duration, MORPH_ACCEL_RAMP_DURATION.
* New feature: Implemented a card, WRITE_ACCEL_RAMP_INFO, to write out the timing of the variable acceleration factor.
* New feature: Implemented a command line argument, INLINE, to run the inline wave model instead of stand-alone.
* New feature: Print header information for Dredge Module to normal location in CMS_DIAG.TXT file.
* Info: Minor changes to screen and diag file output, eliminating duplicate information.
* Info: Minor changes when outputting time information to include number of days.
* Info: Implemented the function equivalents, toUpper and toLower, to the subroutines for converting case.

== Version 5.1.7 (02/25/2019)==
* Bug fix: Read the friction file properly if using ASCII Input.
* Change: Write the Wave Date out in a better format for screen text.
* Bug fix: Resolved a file naming issue when full paths are specified on command line.
* Change: Linux CMake now compiles with OpenMP for parallelization.

== Version 5.1.6 (02/13/2019)=
* Update: Implemented input file changes for Weir and Rubble Mound Jetty structures to match the SMS 13.0+ menu interface design.
* Bug fix: Resolved a few minor issues related to Explicit scheme with parallelization.
* Bug fix: Resolved an issue with hot starting with waves not synchronizing to the right wave condition.
* New feature: EXPLICIT_PRINT_INTERVAL allows user specified interval of time step output.
* Change: Added an additional line of output text to two locations when Morphologic Acceleration Factor used.

== Version 5.1.5 (10/30/2018)==
* New feature: Added Stand-alone wave code that is run when only wave input is specified.  This version is basically identical to last stand-alone code.
** Info: Diagnostic information about stand-alone wave parameters are written to screen and diagnostic file.

== Version 5.1.4 (10/17/2018)==
* Info: Minor version and date changed.  No code changes at all.

== Version 5.1.3 (10/16/2018)==
* Info: Updated Hot Start routines to work with ASCII input and output.  Needs further testing.
* Info: Some work still needs to be done to fully integrate some seldom-used datasets.

== Version 5.1.2 (10/15/2018) [incorporating some developmental features from a non-versioned branch] ==
* New feature: Added card (WRITE_ASCII_INPUT_FILES   ON|OFF) to tell CMS to write out ASCII versions of all input
** Info: All ASCII input files are written to the subfolder "ASCII_Input"
* New feature: Added card (OUTPUT_FILE_TYPE   XMDF|ASCII) to tell CMS to write only ASCII solutions
** Info: All ASCII output/solutions are written to the subfolder "ASCII_Solutions"
** Info: All ASCII Statistics files are written to the subfolder "Statistics"
* Change: Updated internal wave code to latest version, 30 March 2018

== Version 5.1.1 (10/15/2018) == 
* Change: Added in CMS-Wave stand-alone (Mar 2018) for use when only waves are desired
* Info: Investigating reason why old inline wave model doesn't produce same results.

== Version 5.0.4 (04/21/2017) ==
* Compatibility: Reworked several files (mostly explicit-related) for linux compilation.
* Bug fix: Changed variable from REAL to INTEGER as related to explicit boundary conditions.
* Bug fix: Corrected the lack of include statements in some subroutines for compile-time definitions.
* Change: Removed unused statements relating to buffer usage (no performance increase with buffering).
* Change: Moved many WRITE statements with diagnostic output to a separate file.  Will be removed completely later.

== Version 5.0.3 (03/22/2017) ==
* Change: Cleaned up some of the diagnostic output to screen and file.
* Change: Forced location of CMS_DIAG.txt to be in Flow directory if being run, otherwise in the Wave directory.
* Change: Removed debug information in CMS-Wave (GSM TYPE #) and at end of run for dredging module.
* Bug fix: Resolved issues associated with Wave/Flow interpolation in both directions.
* Bug fix: Removed output of 'total.wav' when running with Sea/Swell.  Bug in logic.

== Version 5.0.2 (02/23/2017) ==
* New feature: Dredge module updated to support multiple sediment grain sizes
* Known issue: Avalanching with multi grain sizes is not consistent.  Evaluating for solution.
* Bug fix: Uncommon crash during runs with sediment transport active.  Divide by zero fixed with double precision variable.

== Version 5.0.1 (11/10/2016) == 
* Initial Release of CMS that contains both Explicit and Implicit schemes
* New feature: Supports Explict solution scheme
* New feature: Supports Telescoping files with the Explicit solution scheme
* New feature: Evaporation capability added
* New feature: Dredge module added (single grain size only)

== Version 4.1 52 Beta (09/01/15) ==
* Bug fix: Found related to the screen output for nested water surface elevation and velocity boundary condition which causes an access violation.
* Bug fix: Resolved problem with reading HDF5 solution files from parent ADCIRC simulations.
* New feature: Added the option to specify a maximum bed change per time step to help avoid instability problems. The default value for the maximum bed change in one time step is 0.5 m.
* Bug fix: Found bug in reading none defined spatial coordinate zones as N/A or none integer values. Now any none integer value is set to 0.

== Version 4.1 51 (04/22/15) ==
* Change: Renamed several routines to be more consistent and removed module dependencies from several libraries.
* Change: Removed pure statements from routines and functions since these were causing compiling issues on Garnett.
* Bug Fix: ASCII grid file not being read properly. Modified routine for reading ASCII input file and added format options. 
* Change: Removed wave dissipation override from wave model. 
* Bug fix: Surface roller energy not being interpolated to flow grid when wave volume fluxes are turned off. 
* Bug fix: Found bug in Janssen and Battjes (2007) wave breaking formulation causing wave breaking to be slightly over-predicted.
* Change: Updated wave model to Feb 15 version. 
* Bug fix: Meteorological station wind forcing not being properly applied.

== Version 4.1 50 (12/11/14) ==
* Bug fix: Found problem writing ASCII Cartesian grid file (*_grid.cart). Turned on using OUTPUT_ASCII_GRID ON.
* Bug fix: Found problem applying ramp function to salinity boundary condition. 
* Change: Updated CMS-wave code with 3-2 15May2014 code.
* Bug fix: Found problem in wave dissipation and breaking index when wave reflection is turned on. 
* New feature: Dredging module (beta).
* New feature: Turning on the ASCII output also outputs the boundary ID file (*.bid). This feature is activated using OUTPUT_ASCII_INPUT ON !ON | OFF.  

== Version 4.1 Release 49 (11/14/14) ==
* Bug fix: Found problem reading hot start file when working in a non-local directory.

== Version 4.1 Release 48 (10/31/14) ==
* New Feature: Made gravity an user-adjustable variable. Default value is 9.81 m/s^2. Input format is: GRAVITY 9.8 'm/s^2'
* Bug fix: Found problem reading current velocity initial condition dataset.
* Bug fix: Transport scaling factors not being applied correctly to the Watanabe transport formula.

== Version 4.1 Release 47 (10/10/14) ==
* New feature: Added mean and minimum water levels to the flow statistics, the average and minimum salinity to the statistics, and the mean wave height vector and wave period to the wave statistics.
* New feature: Global statistics are not calculated over and over using the specified start and end times as the duration. 
* New feature: Added option to specify path in XMDF file for initial condition file. For example: INITIAL_CONDITION_FILE  "Initial_Condition.h5" "Datasets/Case05".
* New feature: Added option to specify surface roller model courant number. For example: ROLLER_COURANT_NUMBER 0.3. Default is 0.5.
* Change: Bed layering algorithm changed to prevent bottom layer from becoming too thick. When possible bottom layer is split when merging second and third layer. 
* Change: Minimum active layer thickness separated from minimum inactive layer thickness. Minimum active layer thickness set with ACTIVE_LAYER_MIN_THICKNESS. Minimum inactive layer (2nd and below) set with INACTIVE_LAYER_MIN_THICKNESS.
* Change: Improved screen output and diagnostics
* Change: Minimum wave breaker index set to 0.64. Value first estimated using the Grasmeijer formula and then limited.
* Change: Wave dissipation taken as maximum of wave breaking formula and that estimated from the wave energy field in the surf zone.
* Bug fix: A problem was detected with the OpenMP parallelization in the mixing-length turbulence model which was causing the code stall in some cases. The OpenMP states were removed to fix the problem. 

== Version 4.1 Release 46 (08/21/14) ==
* Change: Changed output format of time in save point output file to scientific notation.
* New feature: Added unit conversions for volume and volume fluxes.
* Change: Modified print format for sediment transport residuals.
* Bug fix: Global sediment balance now considers morphologic scaling factor. Balance is now calculated for each sediment size class.
* Bug fix: Corrected path for CMS-Wave when CMS is run in a subdirectory with no path specified to the CMS-Wave files. The CMS-Wave files are assumed to be in the same directory as CMS-Flow.
* New feature: Added morphology group to save points. The group includes still-water depths and morphology change. 
* Bug fix: Water level offset was not being added to the nested water and nested water level and velocity boundary conditions. 
* New feature: Added ability to skip CMS-Wave output by setting ixmdf = -1 in *.std file. This is useful for long-term simulations when the size of the CMS-Wave can become too large. The wave information can be output instead on the flow grid at much larger time intervals than the steering interval to save space.

== Version 4.1 Release 45 (07/25/14) ==
* Change: Added miles (international) and nautical miles to the units of length in the unit conversion library.
* Bug fix: Corrected a bug in a sediment transport diagnostics routine.
* Maintenance: Changed the names and structure of many of the friction library routines.
* Bug fix: Increased the stack reserve size in the 32bit version to match that of the 64bit version.
* Change: Modified global sediment balance routine to calculate fractional cumulative and net sediment volume balances
* Bug fix: Corrected bug in reading flux data from hot start file causing NaN values.
* Change: Remaining wall clock time calculated using current computational speed
* Change: Bed composition datasets written to *_bedcomp.h5 file instead of *_morph.h5 file

== Version 4.1 Release 44 (05/20/14) ==
* Bug fix: Corrected divide by zero caused by wave period passed from wave model. The problem occured only when for near-zero wave energy spectra and caused NaN values from the first time step.
* Bug fix: Problem fixed when a refinement occurs at the second cell of a boundary cellstrings.
* Change: Cleaned up diagnostics module.

== Version 4.1 Release 43 (05/06/14) ==
* New feature: Option to apply spatial filter to wind velocity fields to remove grid scale oscillations produced by interpolations.
* New feature: Improved output diagnostics for spatially variable wind and atmospheric pressure.
* New feature: Option to output atmospheric pressure gradients.
* New feature: Option to output a scatter set of the wind/pressure field in local coordinates.
* Change: Improved diagnostics for State Plane Coordinate system. Error message indicating invalid zone.
* Change: Improved diagnostics for sediment transport.
* Bug fix: Wind anemometer height correction was not applied to meteorological stations. 
* Bug fix: Atmospheric pressure not being interpolated to ghost cells when using the Oceanweather format causing the spatial gradients to be incorrect.
* Bug fix: Problem reading Oceanweather Atmospheric Pressure file. File was not being opened properly. 
* Change: Improved stability of sediment transport calculations by limiting beta correction factor which is a factor of the bed-load velocity. 
* New feature: Option to specify maximum allowable total-load sediment concentration. The concentration is then limited by the physical concentration limit of the bed concentration.
* Change: Modified convergence criteria for sediment transport avoids early exit.
* Change: Improved diagnostics for NET. The sediment concentration normalized residuals are output when running in debug mode. 

== Version 4.1 Release 42 (04/30/14) ==
* Bug fix: Problem reading wave variable datasets for one-way coupled wave-to-flow simulations. 
* Change: Default smoothing iterations for wave dissipation changed from 0 to 3. It was found that the Lund-CIRP formulas tend to be sensitive to the wave breaking dissipation and generally need some degree of smoothing. 
* New feature: Added diagnostic output for cell ID and maximum sediment avalanching value.

== Version 4.1 Release 41 (04/25/14) ==
* New Feature: Improved error checking when reading input cards and treatment dimensionless variables such as the sediment Shields parameter. 
* New Feature: Added user-specifiable limits to the limits on the hiding and exposure correction in the transport formulas. 
* Bug Fix: Problem specifying the Wu sediment fall velocity and critical shear stress formula. 
* Bug Fix: Incorrect units check for sediment critical shear stress (advanced option within the Sediment Size Class Block).
* Change: Made several notation changes to compile with GNU Fortran using CMake.
* New Feature: Added capability to use arbitrary time units with output time lists and statistcs cards. 
* Bug fix: Intel fortran compiler had an issue with a single-line if statement in the van Rijn sediment transport formula which was causing an early exit and zero sediment transports.
* Bug fix: Problem initializing the bed composition for bed layers 2 and below when bed layer composition is not specified. The default behaviour is to use the bed composition from the deepest specified layer. 
* Change: Removed warning statements when attemping to read hot start datasets from temporary file.
* New Feature: Improved output time display which uses days, hours, minutes, and seconds, instead of just hours.
* Bug Fix: Problem initializing hydrodynamic variables during hot start which was causing instability problems for some cases. 
* New Feature: If the maximum number of sediment iterations (implicit scheme) is not specified than it is automatically determined based on grid size, number of grain sizes, and morphologic scaling factor. 

== Version 4.1 Release 40 (03/10/14) ==
* New feature: Added 40th and 75th percentile diameters as input options of the multiple-sized sediment transport model. 
* Bug fix: Save points not being written out due to initialization problem. 
* New feature: Added output to screen and diagnostic file of the save point settings.
* Bug fix: Corrected problem with sign of negative river flux boundary condition. 
* Change: Slight modifications to wetting and drying scheme. 
* Change: Slight modification to current velocity calculation at open wse boundaries to improve model stability. 
* Bug fix: Spatial smoothing iterations and widths for wse and velocity boundary blocks were treated with a single variable forcing them to have the same value. These have been separated and can now be specified independently. 
* Bug fix: For nested and multiple wse/velocity boundary conditions, previous implementations assumed that the output times were the same for both variables and that the interpolation orders were the same. This has been corrected so that each variable may have its own output times and interpolation order.

== Version 4.1 Revision 39 (02/24/14) ==
* New feature: Turn off Temporal Wave Interpolation when running in steering using the card "TEMPORAL_WAVE_INTERPOLATION".
* Bug fix: Code not writing out SMS Super ASCII files. Bug caused by introduction of preprocessor variables. 
* New feature: Added option to use first order advection scheme. 
* Maintenance: Variable restructuring for faster loop performance.
* Bug fix: Fatal error caused by reading XMDF files without reference time. Error avoided by setting reference time to simulation reference time.
* Bug fix: Problem calculating fraction of suspended sediments when using the Watanabe transport formula.  
* Bug fix: Model output header for suspended concentration files for observation cells even though Quasi-3D is turned off.
* Bug fix: Improper allocation of parent velocity time variable.
* Change: Added a limit of 16 time steps for reduced time steps before increasing the time step.
* Change: Spatial derivative calculation modified so that separate "operators" are used for variables which are a function of wetting and drying (e.g. water levels, current velocities, concentrations, etc.) and variables which are independant of wetting and drying (e.g. bathymetry, atmospheric pressure, etc.). 
* Change: Added OpenMP parallelization to several roller model sections.
* Change: Bathymetry no longer passed from flow model to wave model if sediment transport is turned off.
* Bug fix: Problem determining wetting and drying for hydrodynamic variables at the wave model time step. 
* Change: Bathymetry no longer extrapolated on the wave grid using the bed change. Instead the bathymetry is extrapolated using a weighted distance function of the initial wave grid bathymetry and the extrapolated value.
* Change: Added OpenMP parallelization to several sediment avalanching routine sections. 
* Bug fix: Variable access violation when using a morphologic starting time and multiple grain sizes. 
* Bug fix: Problem in hydrodynamic and sediment transport time-step statistics related to OpenMP corrected by using a firstprivate clause. 
* Bug fix: Current velocity gradients output to incorrect file. 
* Maintenance: Steering routines cleanup. Interpolation now done by library function calls. 
* Beta feature: Added option to pass the bed change from the flow model to the wave model. Specified using: WAVE_BED_ELEVATION  CHANGE !NONE | LAST | CHANGE

== Version 4.1 Revision 38 (01/16/14) ==
* Bug fix: Tidal constituents M1 and L2 had nodal factors
* New feature: Can now read SMS Time-Series Data (*.tsd) files for flux time series boundary condition.
* Change: Specification of non-local horizontal projection not required for nested boundary conditions. This is puts more responsibility on the user to make sure that the parent and child grids have the same projection, but simplifies the model input.
* Bug fix: When nesting a child CMS grid, the model was not properly setting the default names of the parent simulation water surface elevation and current velocity solution datasets. 
* Bug fix: Problem reading SMS XY Series (*.xys) file was corrected caused by FORTRAN allocated statements. 
* Change: Improved diagnostics for SMS XYS and TSD files, and meteorological library.
* Maintenance: Made variable name changes to avoid conflict with merged explicit/implicit CMS code. 
* Bug fix: Problem in reading mmHg and inHg units for pressure do due case sensitivity. 
* Bug fix: Problem extrapolating using linear piecewise interpolation with only two data points.  
* Bug fix: Problem in reading Standard Meteorological ASCII files (*.txt) when launching the executable using a drag-and-drop option. 
* New Feature: Added capability to export Time Series Data files containing boundary wse and velocity forcing when nesting CMS.
* Bug fix: Bed-slope coefficient in hydrodynamic equations was not considered in momentum equations even when turned on. 
* Beta Feature: Incorporation of cross-diffusion terms in implicit time-stepping scheme as a source term. 
* Beta Feature: Added ICCGSTAB solver. 
* Change: Begin and end banners changed to display decimal seconds. 
* Bug fix: Corrected OpenMP statement in spatial derivatives library causing the serialization of two loops.
* Beta feature: Added optional second-order finite-difference approximation to spatial derivatives for nonuniform Cartesian grids.
* Bug fix: Corrected bug in subroutines related to deferred corrections for higher-order HLPA and Gamma-family advection schemes on skewed grids.

== Version 4.1 Revision 37 (11/21/13) ==
* Bug fix - Incorrect error message when reading local horizontal projection system
* Bug fix - XMDF library causing error en closing temporary file temp.h5. Code added so that it does not crash the model but the temporary file is not deleted. 
* Bug fix - Latitude dataset was not being read in properly.
* Bug fix - When hot starting a simulation, if the initial condition for the salinity did not match the boundary conditions, this led to an inconsistency at the boundary. The salinity boundary condition is now ramped from the initial condition to the specified time series values using the same ramp function as in the hydrodynamics.
* New feature - Added more optional strings units: seconds, radians, and degrees.
* New feature - Added preprocessor definition for diagnostic mode
* New feature - Improved diagnostics message 
* Bug fix - Bed layer thickness not being properly read in during hot starts.
* Bug fix - When hot starting a simulation with sediment transport, the water depths at the ghost cells was not being set properly causing instabilities for only the first time step.  

== Version 4.1 Revision 36 (10/28/13) ==
* New feature - Added timers (watches) to profile the code, with C preprocessor definitions, and unit tests.

== Version 4.1 Revision 35 (10/10/13) ==
* New feature - Added C preprocessor definitions for XMDF source code at compilation time. We be used to include/exclude other options such as profiling in the future.

== Version 4.1 Revision 34 (09/19/13) ==
* New feature - Added capability to specify save points using a block structure
* New feature - Added capability to read wind speed and direction from SMS xys files
* Bug fix - The temporal scheme was not being set to first order after a time-step reduction. 

== Version 4.1 Revision 33 (09/16/13) ==
* Bug fix - Fixed divide by zero in combined wave-current Soulsby-van Rijn subroutines
* Bug fix - Moved definition of default value for variable nhard to initialization routine sed_default

== Version 4.1 Revision 32 (09/12/13) ==
* Bug fix - Hard bottom datasets not being read in properly
* Bug fix - Corrected bug related to beta feature allowing the bed composition to change while not the bed elevation

== Version 4.1 Revision 31 (09/11/13) ==
* New feature - New open wse boundary velocity condition based on approximate momentum equation and damping factor
* Bug fix - Incorrect warning message displayed when the flux boundary inflow angle is specified outwards of the domain. The actual flux at the boundary was correct and there was a bug in the code which checks the boundary flux.

== Version 4.1 Revision 30 (08/30/13) ==
* Update - Added new version of wave model 20 May 2013
* Update - Added beta feature to set start of bed composition and bed change separately. This allows the user to estimate the equilibrium bed composition for a user-specified equilibrium bathymetry

== Version 4.1 Revision 29 (07/12/13) ==
* Bug fix - CMS-Wave was allowing for the user to select a number of frequencies which was larger than then hardcoded value 

== Version 4.1 Revision 28 (06/18/13) ==
* Bug fix - Morphology start time not set properly when using advanced cards
* Bug fix - Unidentified memory problem with OpenMP parallelization in avalanching subroutine. Parallelization removed until the problem can be fixed.
* Bug fix - Found bug in avalanching for telescoping grids leading to unsymmetrical results.

== Version 4.1 Revision 27 (06/17/13) ==
* Bug fix - Flux correction was being applied between dry and dry cells
* Bug fix - Found bug in searching for overlapping points in subroutine intpcoef_cart2tel
* New feature - Added atmospheric pressure and stokes velocity to water level adjustment at boundary conditions.
* Change - Modified boundary treatment for pressure equation using momentum interpolation. Changes done in subroutine prescorbnd and improved the boundary for field cases. The change also simplified the code.
* Bug fix - Fixed bug related to predicted wave water level for constant water level boundary condition (beta). 
* Bug fix - Fixed bug related to the current velocity specification at open water level boundaries. The current velocity is calcualted using an extrapolated momentum interpolation. The bug fix has resolved instabilities causing jets to form at the boundaries. 
* Bug fix - Reference time not being set properly for wave vector and radiation stresss output in XMDF file
* New feature - Added capability to use SMS Time Series Data files for wind input curve specification.
* Change - Improved diagnostics. Now outputs diagnostic super ascii files in the case of a crash. In addition more variables are displayed when an instability is detected. 
* New features - Quasi3D formulation (beta).Completed the LSTF and LIP11D-1B test cases. Added capability to output vertical profiles of the horizontal current velocity at observation stations and global datasets for the current velocity at different sigma levels.
* New feature - Option for bottom boundary layer streaming added as a shear stress at the bottom in the direction of waves (beta)    
* Change - Consolidated math routines into a library.
* Change - Modified water level boundary condition for flux boundaries to use an extrapolated water level.
* Maintenance - Cleaned up bottom and wall friction library. Renamed several subroutines and variables. 
* New feature - Addes routines for interpolation from polygonal to Cartesian grids and Cartesian to scattered points.
* Modification - Changed interpolation coefficient file format to only save non-zero coefficients. 
* New features - Added option to output: cell area, connectivity, total flux velocity vector and magnitude, wave flux velocity vector and magnitude, total water depth, water level pressure, water level gradients, current velocity gradients, streamwise curvature, and roller dissipation.
 
== Version 4.1 Revision 26 (05/30/13) ==
* Bug fix - Fixed typos in vertical units and datum definition (beta)
* Bug fix - Ramp function was not being properly assigned to the bottom orbital velocities

== Version 4.1 Revision 25 (05/07/13) ==
* Bug fix - Resolved bug related to harmonic boundary specification. 
* Fixed bug - Fixed bug related to the boundary extrapolation at fitted boundaries (alpha)
* New feature - Added cards to read wind stress and atmospheric pressure fields
* Change - Salinity horizontal mixing coefficient divided by Schmidt number
* Change - Changed output format for super ASCII files
* New feature - Super ASCII files now include residuals, total water depth, pressure, and pressure correction
* Bug fix - A problem was found with the parallelization of HLPA scheme and nonuniform Cartesian grids using the Bump test case due to the NOWAIT clause. 
* Bug fix - Fixed bug in roller module related to changed common block for CMS-Wave files. The bug was introduced with a new version of the wave model (10Oct2011) and was causing the wave files to be mixed up. The wave XMDF output file was being saved with the name wind.dat and in subsequent runs the file was being read in as a wind file causing an error.
* Bug fix - Fixed bug related to advanced constant roughness height specification
* Bug fix - Fixed bug related to improper initialization of error diagnostic variable ierr in unit conversion library. Problem only occured when running in debug mode. 
* Bug fix - Fixed bug related to dataset card read which was overwriting the relative path. 

== Version 4.1 Revision 24 (05/07/13) ==
* Bug fix - Fixed devide by zero problem in SIP solver leading to NaN's
* Change - Changed sediment bed initialization so that the input first layer is taken as the mixing layer
* New feature  - In the case the user specifies the water density and temperature the model now solves the equation of state for the water salinity using the secant method to estimate the water salinity which is then used for the kinematic viscosity calculation
* Bug fix - Fixed bug related to grain size calculation for older 3.75 where only the d50 dataset is given without a transport grain size.
* New feature - Added code for user to modify the maximum matrix solver iterations for the sediment concentration using the card SEDIMENT_CONCENTRATION_ITERATIONS. 
* Bug fix - Fixed bug related to harmonic blocks
* New feature - Added option to turn on or off the wave and wind adjusted water level
* Change - Restructured advanced boundary nesting variables to that parent simulation is stored separately.

== Version 4.1 Revision 23 (04/25/13) ==
* Maintenance - Cleaned up salinity boundary condition initialization
* Bug fix - Fixed bug related to implicit salinity transport boundary conditions
* Bug fix - Fixed bug related to implicit salinity transport normalized residual initialization
* Bug fix - Fixed bugs related to new spatially variable wind formats. Fleet format and UVP formats have been verified.
* Change - Changed save points so that the cell id is always calculated using the coordinates
   Maintenance - Cleaned up met station routines and verified met station input. Enhanced met station diagnostic output.

== Version 4.1 Revision 22 (04/23/13) ==
* Bug fix - Units of minutes were not being read in correctly within the unit conversion library. 
* Bug fix - Fixed bug related to single size sediment transport 
* Bug fix - Added parenthesis before viskfl in subroutine. Error only an issue for constant sediment mixing coefficient (advanced alpha feature). 

== Version 4.1 Revision 21 (04/19/13) ==
* Change - Simplified the mixing layer thickness calculation so that it is no longer a function of the deposition thickness
* Change - Changed interpolation of cell face bed load transport rate in bed slope term to harmonic mean
* New feature - Added parallelization to btklogexp and btklogrouse subroutines
* Change - Added new hardbottom treatment to limit concentration capacity
* New feature - Added parallelization to sedcapac_soulsby subroutine
* Bug fix - Fixed a bug related to the second order temporal scheme in which the variables from n-2 time step were not being updated properly for the first time step during variable time steps. 
* Change - The second order scheme is only applied if the time step does not change.
* Bug fix - Logical variables for single tidal constituent boundary were not being initialized properly causing the model to not run on certain computers

== Version 4.1 Revision 20 (04/10/13) ==
* Change - Modified one-cell-wide-channel wetting and drying criteria to allow flow in deep and narrow channels

== Version 4.1 Revision 19 (04/09/13) ==
* Change - Added additional debug variables in the diagnostic solution file *_diag.h5

== Version 4.1 Revision 18 (04/08/13) ==
* Bug fix - Fixed bug related to path specification for ICoords and JCoords. Bug only occurred for one known grid but has been fixed.
* Bug fix - Fixed bug in reading the cross-shore boundary condition. 
* New feature - Added cards to specify convergence thresholds for implicit solution scheme
* Change - Moved precipitation and evaporation contribution to source term into the static source term for the implicit scheme

== Version 4.1 Revision 17 (04/06/13)
* Bug fix - Changed allocation size for vdata and udata arrays in nesting structure variable NH_str
* Change - Added internal closed boundary condition to improve the pressure field solution for the implicit scheme
* Bug fix - Added temporary variable to avoid changing the time series step variable in subroutine tidevalue within the steering module

== Version 4.1 Revision 16 (03/15/2013) ==
* Bug fix - Moved initialization of nestin1, nestin2, and iwave in CMS-Wave to the beggining of the file
* Change - Added input card for maximum current velocity, and water level values for convergence checking

== Version 4.1 Revision 15 (03/14/2013) ==
* Bug fix - Changed input units for single size sediment transport cards from m to mm
* Change - Removed slope limiter from pressure gradient term. This was causing convergence problems. 
* Bug fix - Changed if statement in Soulsby-van Rijn transport formula for wave conditions (.not.cmswave) tp (noptset>=3)

== Version 4.1 Revision 14 (03/12/2013) ==
* Bug fix - Removed extra ramp from boundary concentration in subroutine sedbnd_eval

== Version 4.1 Revision 13 (03/04/2013) ==
* Bug fix - Added interpolation of gaps (dry values) for internal nested boundary conditions

== Version 4.1 Revision 12 (03/01/2013) ==
* Bug fix - Fixed critical shear velocity correction for SVR and VR transport formulas
* Bug fix - Fixed path bug in CMS-Wave
* Bug fix - Fixed path bug for boundary conditions
* Change - Increased default wave bottom friction factor to 0.65 from 0.5 based on comparisons to other models

== Version 4.1 Revision 11 (02/09/2013) ==
* Bug fix - Fixed bug in wave friction factor calculation for Data2 and Data13 (only for smooth conditions)
* Maintenance - Cleaned up the friction library
* New feature - Added bedform routines based on Soulsby (2005)

== Version 4.1 Revision 10 (01/17/2013) ==
* New feature - Least squares interpolation
* Maintenance - Moved structures code to separate files.
* Maintenance - Cleaned up wetting and drying subroutine. Added parallelization to onecellwidechannels special wetting/drying case loop. Moved distance to wall calculation to eddy viscosity module and added parallelization. Added parallelization to flpond.
* Bug Fix - Revised deferred correction subroutines and fixed OpenMP race condition
* Maintenance - Changed code structure so that component variable definitions are in fortran modules in separate files named *_def.F90 (e.g. sed_def.F90), and component libraries are in separate files named *_lib.F90 (e.g. sed_lib.F90). Also separated source into folders so that is better organized.
* Change - Revised new block structures for boundary specification
* New feature - Added OpenMP parallelization to sediment avalanching subroutine
* New features - Added time step statistics to sediment transport
* Change - Moved water density from module to input in all friction subroutines.
* Change - Cell-face water depth is not calcuated from cell face water elevation, and bottom elevation instead of interpolating the water depth between cells. The new approach more efficient since the bed elevation only changes once per time step, and the water level gradients are calculated anyway.
* Bug fix - Fixed bug related to bed slope correction in Lund-CIRP and Soulsby-Van Rijn transport equations.
* Bug fix - Fixed bug relates to simplified log-law wall friction formulation
* Bug fix - Corrected if statement in iterative log-law wall friction formulation 
* Change -  Changed wall friction formulation to use wall friction coefficient instead of gamma

== Vesrion 4.1 Revision 5 ==
* Change - Added parallelization to hardbottom loop in sed_imp

== Version 4.1 Revision 4 ==
* New feature - Added grid quality indicators to simulation statistics module
* New feature - Polyhedral code complete and under testing

== Version 4.1 Revision 3 (11/30/2012) ==
* Maintenance - Added polygonal mesh support for reading *.2dm files and started inserting unstructured code from polygonal version
* New feature - Added subroutines for reading card with scalars and booleans
* New features - Added unit conversion subroutines
* Maintenance - Cleaned up boundary initialization subroutine

== Version 4.1 Revision 2.5 (11/26/2012) ==
* Change - Separated smoothing iterations for wave breaking index and wave breaking dissipation
* Change - Changed default smoothing iteration numbers for wave variables
* Change - Added steering variables for controling flow-to-wave current velocity and bed elevations values
* Maintenance - Cleaned up setwave subroutine and added OpenMP parallelization
* Bug fix - Extrapolation distances were being reset to zero when reading interpolation files
* New feature - Added SIP and ICCG solvers for structured grids

== Version 4.1 Revision 2 ==
* Maintenance - Added parallelization to several sections of code
* Bug fix - Removed race condition in deferred correction subroutines
* Maintenance - Moved boundary condition specification to subroutines
* New feature - Different formulations for spatial gradients now available. 
* Bug fix - Fixed boundary condition bug (introduced in V4.1)

== Version 4.1 Revision 1 ==
* Maintenance - Added parallelization to several sections of code
* Maintenance - Combined coefficient assembly for u and v momentum equations
* Maintenance - Combined coefficient and source term calculations for pressure correction equation
* New feature - Added meteorological stations
* New feature - Added general input cards for scalars and booleans

== Version 4.1 ==
* New feature - Nested water level boundary condition in a larger CMS or ADCIRC grid
* New feature - Nested water level and velocity boundary condition in a larger CMS or ADCIRC grid
* Change -  Code for calculation of interpolation coefficients and index made into subroutines
* New feature - Added basic linear algebra subroutines with loop unrolling and OpenMP for faster computations.
* New feature - Added spatial water level gradient to tidal, and single water level boundary conditions
* New feature - Added mean water level to tidal/harmonic boundary condition
* New feature - Block structure input for tidal/harmonic boundary conditions
* New feature - Extended the range of the yearly nodal factor and equilibrium database to 1850-2050.
* Maintenance - Cleaned up sparse matrix module and added several comments.

== Version 4.0 Revision 14 ==
* Bug fix - Fixed cdWndareap calculation for Lagrangian winds.
* Bug fix - Variable sqrttwo not defined properly.
* Change - Initial guess of roller energy is set to the previous time step instead of zero and converges much faster.
* New feature - Added output for fraction of broken waves and wave dissipation coefficient.
* Change - Modified incipient wave breaking.

== Version 4.0 Revision 13 ==
* Change - Modified representative bottom orbital velocity Uwr calculation using parametric method based on JOHNSWAP spectrum
* Change - Modified bottom friction calculation to that Uwr is used instead of Uw for all of the bottom friction friction formulations except the quadratic one
* New feature - Added card MORPH_START_TIME which replaces CALC_MORPH_DURING_RAMP
* New feature - Added GM79 and DSK88 bed friction formulations
* Maintenance - Verified Soulsby-van Rijn transport formula with example case from Soulsby 1997
* Maintenance - Verified GM79 bed friction formulation using example from Soulsby 1997
* New feature - Added bed-correction to Soulsby-van Rijn transport formula 
* Change - Simplified wall friction formulation to avoid iterative solution
* New features - Added curvature to hydrodynamic simulation statistics

== Version 4.0 Revision 12 ==
* Bug fix - Hot start failed to write to hotstart file after starting from hotstart file.
* New Feature - Save Point files instead of Observation Cells.  Ensured this works with Hot Starts
* Change - Modified HYDRO_TIMESTEP, DURATION_RUN, DURATION_RAMP to be able to read time units.

== Version 4.0 Revision 11 ==
* Change - Set default output file names for each group to group names instead of solution file
* Change - Added cards to turn on/off specific output variables
* Bug fix - Fixed ascii output of current velocity
* Maintenance - Cleaned up prestart, read_card_file, and others    
* New Feature -  Added Quasi-3D module to calculate 3D dispersion terms  
* Change - Modified cards for wave mass flux velocity and roller flux

== Version 4.0 Revision 10 ==
* Change - Removed coefficient calculation for v-momentum equation, except at boundary cells
* Change - Added static source terms for momentum and salinity
* New Feature - Added second-order temporal scheme
* New Feature - Added GAMMA, CUBISTA, ALSMART, and HOAB advection schemes
* Change - Optimized deferred corrections
* Change - Improved output display screen
* New Features - Added rubble mound, culverts, and weirs from Weiming's code.

== Version 4.0 Revision 9 ==
* Bug fix - Removed average current velocity output in sim_stats causing error
* Change - Units of sediment transport from m^2/s to kg/m/s
* Maintenance - Removed unused variable struct from code
* Change - Removed positive and negative sediment transport rates from statistics
* Change - Added statistics output for maximum sediment transport rates
* Change - Added statistics output for total-load concentration norm residuals

== Version 4.0 Revision 8 ==
* Change - Changed variable time step settings (more conservative)
* Bug fix - Reordered convergence if statements in CMS_Flow
* Bug fix - Removed mean current velocities from simulation statistics (not used currently)
* Bug fix - Changed sign of "Stokes forcing" term in momentum equation (negligible effect)
* Bug fix - Removed wccg reference in roller and replaced with wccg2 (wccg2 created by Alex)
* Bug fix - Added reinitialization of tmultw in recalculate (to properly recover from crashes)
* Change - Added cdflux1 variable to be used in recalculate and update (more consistent)
* Bux fixes - Several bug fixes reported bu Earl Hayter for CMS-Wave (EJH)
* Bug fix -  Added code to not allow vis=0.0 at dry cells (EJH)
* Maintenance - Removed old code for reading NCCHE type input ASCII files
* Bug fix - Divide by zero in sed_betat for dry cells due to zero viscosity
* Change - Reintroduced the wave ASCII output for use with PTM

== Version 4.0 Revision 7 ==
* Bug Fix - Replaced wucapac transport formula

== Version 4.0 Revision 6 ==
* Bug Fix - Bug fix for multiple XMDF output files. Files are deleted before for coldstarts.
* Bug Fix - Wind linear interpolation coefficient changed to avoid extrapolation.
* Bug Fix - Changed observation station output format to F9.4 so that dry cells write -999 for water elevation
* Bug Fix - Increased tolerance for output in check_time_list to avoid skipping output times
* Change - Increased interval for sediment transport screen print from 5 to 10

== Version 4.0 Revision 5 ==
* New feature - Hybrid solver using both GMRES and Gauss-Seidel
* New feature - Hydro statistics are used to check for model divergence at every outer loop iteration.

== Version 4.0 Revision 4 ==
* New feature - Multiple XMDF output files for each variable group.

== Version 4.0 Revision 3 ==
* New feature - Display time step statistics.

== Version 4.0 Revision 2 ==
* Change - Variable time stepping algorithm. 

== Version 4.0 Revision 1 ==
    Bug Fix - NaN wind drag coefficient produced for zero wind speed.