Changelog
Version 5.0.2
This release was created to auto-publish the binaries for watem-sedem on GitHub.
For windows: watem_sedem.exe For linux: watem_sedem For debian/ubuntu linux also an apt package is provided.
Version 5.0.1
This release contains some small updates to the documentation and all model extensions are disabled by default in the model code. The user has to enable them in the ini-file.
Version 5.0.0
This release contains a name change from CN-WS to WaTEM/SEDEM (in code watem_sedem) in order to follow the initial aim of the KU Leuven for developing WaTEM/SEDEM: CN was added in later versions. The CN-module is now refactored to an extension. With this release, a distinction is made between the basic WaTEM/SEDEM model and the extensions (CN, buffers, forced routing, etc..).
In this version following changes were created:
Restructuring the *.ini-file:
The section “User Choices” was split in two new sections:
Options
Extensions
The section “Variables” was split in two new sections:
Parameters
Parameters Extensions
The section “Output maps” was renamed to “Output”
The option “Only WS” was renamed to “Curve Number”
The executable/build cn_ws was renamed to watem_sedem
The naming in the folder structure was updated from cn_ws to watem_sedem
Multiple additions/corrections to the documentation
Version 4.2.3
Multiple additions/corrections to the documentation.
Fix of routing to sewers-bug introduced in CN in version 4.2.0 (see 6735f759b5029c63a4531e96aa823484722e48b3). Discharge observed in segments is now correct again.
Add GitHub actions.
Version 4.2.2
Implements option that allows only cardinal routing to the river. This option is usefull to avoid ordinal routing between two grass strip pixels. As a consequency, the cardinal routing is a way to maximize impact of grass strip pixels in the simulation.
Version 4.2.1
Add parcel connectivity grasstrips as a parameter.
Write ‘routing_missing.txt’ only when ‘write routing table’ is enabled in the ini-file.
Disable option to calculate R-factor in CN-WS, referring to the R-factor python package: https://watem-sedem.github.io/rfactor/index.html.
Enable install with cn_ws with conda.
Several documentation updates, additions and clarifications.
Refactoring code
version 4.2.0
Extending documentation
Bugfix routing: if both targets have a different land cover than the source pixel, and the highest target is a gras strip, than route to the other target.
Renamed model binaries and folder structure:
console version: CN_WSModel > cn_ws
gui version: CN_WS > cn_ws_gui
LT version: CN_WS_LongTerm > cn_ws_gui_LT
Moved logic of lpr file to a separate .pas file
Remove spurious warnings when building the binary
remove VHA tag in files:
The option “Output per VHA river segment” became “Output per river segment”
Discharge_VHA.txt was renamed to Discharge_segments.txt
Total sediment VHA.txt was renamed to Total sediment segments.txt
Cumulative sediment VHA.txt was renamed to Cumulative sediment segments.txt
Sediment_VHA.txt was renamed to Sediment_segments.txt
Sediment concentration_VHA.txt was renamed to Sediment concentration segments.txt
Clay content sediment VHA.txt was renamed to Clay content sediment segments.txt
Enable the output per segment option by default when river routing is used
Make content of Total Sediment.txt dependent on used options in ini-file
- Correct tillage erosion:
sin –> tan in formula Van Oost et al., (2000)
use the same routing for tillage erosion as in the water erosion
Add SEDTIL_IN and SEDTIL_OUT-files to output
Bugfix: do not overwrite determined routing for buffer pixels with standard routing
Bugfix: reduce uparea from buffer outlet pixel when Buffer reduce area is enabled
Add file rfactor.txt to modeloutput when the rfactor is calculated from a rainfall timeseries
adjusting sewer routing CN to current sewer routing WS.
Enable option to input precomputed R-factor for a single event in CN-WS.
Make CN-WS compatible with SAGA-grids (new option).
Make integer rasters int32-compatible (requires Freepascal 3.2): this opens the option to run larger catchments (with int16, the number of parcels are limited to 32768).
Correct calculation S-factor Desmet (use tan in stead of arctan)
version 4.1.0
This version was delivered to VMM at the end of the NEMO project. It includes some small improvements and a fix for a bug discovered by Jan Coppens (Thanks!)
Allow relative paths in ini-file
Simplify logic in CN_WSmodel.lpr
Make bufferparameters ‘Volume’, ‘Height Dam’, ‘Height opening’, ‘Opening area’, ‘Discharge coefficient’, ‘Width dam’ only mandatory in ini-file when using the CN module
Bugfix: Set Target2 to -99 when Part2 is 0 (as is done for Target1)
Added TC formula of Verstraeten et al 2007 as an option
Bugfix: Renaming SModel Desmet1996 to McCool1987
Extending documentation
version 4.0.0
This version includes several changes needed for CNWS to be compatible with the NEMO model of VMM (see https://www.vmm.be/water/kwaliteit-waterlopen/waterkwaliteitsmodellen/nemo_tw.pdf/@@download/file/NEMO_TW.pdf )
Change type of parcel raster to integer
Overwrite the system decimal separator to ‘.’ cn_ws can now be used for both a float comma and point decimal system
Add documentation with sphinx
Don’t write Total Sediment.txt when calibrating
Bugfix: read outlet raster when ‘Manual outlet selection’ is enabled
Improving memory management (releasing memory of rasters when they are not used anymore)
Change buffer extension ID to buffer_id + 16384
Routing: do not jump to border pixels of raster in findlower function
Removing ‘Sediment trapped in open water’ from Total sediment.txt
Round sewer input in Total sediment.txt to two decimals
Cleaning code: removing initialisation of unused variables
Determine outlet after routing when ‘Manual outlet selection’ is enabled
Refactoring: rename variable FLUX_IN to UpArea
Use sewermap to reduce uparea
Add optional LS correction factor
version 3.0.0
do not route sewers to rivers
reduce sediout of sewer pixel with fraction defined in sewermap
write capacity to raster
removing outputfile ‘Sewer output sediment.txt’
Adding ‘Sediment entering sewer system’ to Total sediment.txt
Bugfix: typo in variable name (2nd target vs 1st target)
If Target is -99 set distance to 0 in routing table
Change option ‘Include tillage’ to ‘Include tillage direction’
Make calculation of tillage erosion optional
Change Fluxout from RRaster to single value
Ring checking
Use function to find lower pixel
Update routing algorithm (inverse routing)
Allow to stop routing algorithm in rivers and outlets
After using ‘dedicated’ routing in ditch, dam or buffer make sure to use ‘normal’ routing algorithm
Reduce upstream area with trapping efficiency of buffer
version 2.1.0
Add option ‘Only Routing’
Make routing table an optional output
Endtime model variable is now only mandatory in ini-file when needed
Add check to make sure number of river segments in rasters and files are the same
Set the ktc value of a dam to ktc_low instead of 7
Flow to neighbour river cell even if that cell is higher as the origin
Make sure that sediment transport uses the same routing as upstream area
Fix calculation of distance between source and target (wrong in peculiar cases)
Use river direction as input for river routing
Write sewer export to rasterfile
Add files to build a package on debian linux (for deploy on notebook server)
version 2.0.0
Version used for calibration with an updated version of the digital elevation model of Flanders (DHMVII).
Add calculation of cumulative river sediment
Don’t write empty routing lines in routing.txt, use tab separation
Use proportion for upstream edges
version 1.4.0
This version was made for the third steering group of the ‘calibration DHMVII project’.
remove name of ini-file from files-section in ini-file
add option ‘Buffer reduce area’
enabling range checking to avoid errors
PTEF is changed to base-100 (percentage) while it used to be 1-based
Remove sediment trapping in open water (-5 in parcel map) pixels and assign ktc 9999 to those pixels
bugfix to prevent out of range when calculating adjusted slope
Change ktc values to float (previously integer) and adapt calibration accordingly
Skip ktc low values higher than ktc high in calibration mode
Adding Force Routing option
Don’t route cells without lower cells to themselves
Improved error message when input directory is missing
Add River Routing option
Change default value of Create ktc map to True
Refactoring code
Numerous model simplifications
version 1.3.0
This version was made for the second steering group of the ‘calibration DHMVII project’.
Use adjusted slope calculation in LS calculation
Restructure code so slope and aspect are calculated once, instead of twice
Add -9999 as no data value in all output rasters
Improve memory allocation for rasters
Refactoring code to read idrisi rasters
version 1.2.0
This version was made for the first steering group of the ‘calibration DHVMII project’.
raise exception when ini-file does not exist
make key words in ini-file of LT version the same as in other version
bugfix: distance calculation near buffers
write routing table as output of console version
raise exception if no outlet is present in outletmap
add search radius to ini-file (make it a variable, not a constant)
add calibration method (loop over all combinations of ktc low and ktc high in a certain amount of steps and write output to calibration.txt)
add functionality to use different L (Desmet1996_Vanoost2003 and Desmet1996_McCool) and S models (Desmet1996, Nearing1997)
return non-zero on unsuccessful exit
refactoring and cleaning code
amount of sediment at outlet is now calculated as sum of incoming sediment in the segment that contains the outlet
version 1.1.0
three code bases (gui, console and long-term version) are merged to a common code base
add function to write routing table
correct distance calculation
improved exception handling
cleaning code
Version 1.0.0
Initial version of CN-WS, as developed at KU Leuven.