2CSalt

Software

Support

Contact

Phone: 1300 5 WATER (1300 5 92837)
Email: support@ewater.org.au

eWater
Innovation Centre
University of Canberra
ACT 2601
www.ewater.org.au

History & Credit

Version history

Version 1.0.8.1

Released: Jul 15, 2009
Licence Date: Dec 9, 2005

This is a minor release.

Bug Fixes

Fixed an issue with an error message being thrown for an invalid climate data field in the generate HRU map module.

Software Enhancements

None in this release

Science Changes

No Science changes in this release

Speed and memory improvements

The Water Balance Module is now threaded and should run at close to n-times (where n is the number of CPU cores on system) faster.

Issues

None in this release.

Version 1.0.6b

Released: Aug 28, 2008
Licence Date: Dec 9, 2005

This is a minor release to reset the activation code.

Version 1.0.5b

Released: Feb 27, 2008
Licence Date: Dec 9, 2005

This is a minor release to reset the activation code.

Version 1.0.4b

Released: Aug 21, 2007
Licence Date: Dec 9, 2005

This is a minor release to reset the activation code.

Version 1.0.3b

Released: May 9, 2007
Licence Date: Dec 9, 2005

This should be the last minor release before public release.

Bug Fixes

Software Enhancements

Science Changes

Several as yet undocumented science changes

Speed and memory improvements

Issues

The move to .NET 2.0 has introduced speed improvements to most of the Raster and TimeSeries Based analysis but slowed most Vector based analysis. The slowdown is due to the poor performance of the ?.Contains()? method in System.Collections when they contain copy by reference objects. We are hopeful this will be fixed by Microsoft in the near future. *NOTE* this issue only seems to affect Intel Pentium 4 based architectures

Version 1.0.1b

Released: Nov 7, 2006
Licence Date: Dec 9, 2005

This should be the last major release before public release.

Bug Fixes

Addressed an issue where single cell sub-catchments with valid nodes were not assigned a value for length and height.

Software Enhancements

The 'remove noise' Class now can be set to protect Sub-Catchments that have a valid node. These Sub-Catchments will not be removed if this option (available in the User Options Dialog 'File > User Options') is set to true.

The 2CSalt software has been migrated to the .NET 2.0 framework.

The ShapeFileLoader no longer uses the Microsoft?s data access components so should be more robust.

There have several major changes to the use of the water balance flux DataCubes:
1) 2CSalt now requires the HRU to be loaded before the water balance DataCubes. This allows 2CSalt to only load into memory the data actually required to create a scenario. Consequently very large disk based DataCube libraries can now be used with no impact on memory.
2) After loading the water balance DataCube?s the subset actually present in the HRU map and resident in memory can now be save to disk as new files.
3) 2CSalt now uses a binary DataCube format in addition to the ASCII one. The binary DataCube files can significantly speed up the loading of large water balance DataCubes into the 2CSalt application.

A new analysis tab ?Water Balance Module? has been added to the 2CSalt interface. This module allows the water balance DataCube?s to be build within the 2CSalt application using a version of the PERFECT model ported to the TIME framework.

Science Changes

No Science changes in this release

Speed and memory improvements

The support for large Raster datasets has been significantly improved through a number of changes. These changes can be set in the User Options Dialog ?File > User Options?

All Raster states now invisibly support single precision. This will almost halve the memory usage of most modules with no loss in function or degradation of results.

A new disk based Raster State has been added called DiskTriLine that is much faster than the previous disk based Raster support states and has the added bonus of using typically less than 1 mb of memory per Raster. It can be activated in the User Options dialog. By default a Raster State will be disk based if it is larger than 30 million cells.

Complete re-write of the Raster spatial apportionment Class. This Class is used in the terrain analysis module. This new Class uses very little memory and is substantially faster than the previous class.

Complete re-write of the noise removal Class used in the terrain analysis module. It is now significantly faster and uses a fraction of the memory of the old Class.

The Node apportioning Method has been re-written to improve speed.

The ellipse fitting Class has been tweaked to improve speed.

The Main Form can now be maximised on certain modules

Issues

Version 1.0.0b

Released: Dec 9, 2005
Licence Date: Dec 9, 2005

Beta Release Version of 2CSalt

Version 0.9.8

Released: Jul 19, 2005
Licence Date: Mar 16, 2005

Version 0.9.7

Released: Jul 8, 2005
Licence Date: Mar 16, 2005

What?s new in this release?

1 - Fixed salt calculation bug that allowed the runoff salt store to go negative.

2 ? Fixed a bug that stopped a GRU map from running if a GRU contained no alluvial store.

3 ? Added a re-sampling option to the ?GRU Options? menu. This will be useful for working with large DEMs. Currently it uses a nearest neighbour technique.

4 ? Added a new option to the terrain analysis module that Anna developed at the TIME training course that will extend the produced sub-catchment map to the size of the input DEM using an onion skin approach. This approach does not take hydrology into account so use with caution. It can also take a significant amount of time to run on large/high resolution DEMs.

5 ? I?ve removed the slider bar from the sub-catchment threshold as it was causing rounding errors.

6 ? I?ve added some more options to the ?User Options? form. One allows you to turn of the progress form displayed by the PolygonTools at various stages of the applications operation. Under data types there are several options that allow the use of very large Rasters by either using memory compression or caching the Raster to disk. The modeller can set a threshold for the number of cells a Raster must have before it is cached to disk. A more detailed explanation will be included in the next version of the documentation.

7 ? The current version can now save a Raster to all the file formats supported by the TIME libraries.

8 ? Added support for Surfer ASCII GRID IO.

9 - Fixed a small but that under certain circumstances may have made the save Depressionless DEM button disabled.

10 - Added support for the optional use of a vector based analysis mask.

Version 0.9.5

Released: Jun 1, 2005
Licence Date: Mar 16, 2005

What's new in this Version?

1-You can now merge polygons adjacent on the same stream path

2-The program now checks that the network topology is formed properly during the terrain analysis step and bails if it detects more than one outlet from the catchment.

3-You can now view the network topology instead of the stream raster. To do this click the option in the 'GRU Options' menu.

4-You can switch off the stream network checking in the 'GRU Options' menu. Doing this allows merging and deleting of any sub-catchments. It also allows sub-catchment?s maps to be formed with multiple outlets.
Note: By switching of the stream network data checking the output files may no longer be compatible as input to E2.

5-The subCatchmentSplitter now rebuilds the stream network topology 6-Runoff is now limited to 10000 mg/l for each runoff event.

Version 0.9.4

Released: May 25, 2005
Licence Date: Mar 16, 2005

What?s new in this version?

1 ? I?ve added a slider bar and percentage box to help with selecting a starting value to delineate the DEM into sub-catchments.

2 ? I now track the network topology in the terrain analysis section. This allows me to control how the catchment is modified. For example you can now only delete sub-catchments that have no up stream sub-catchment. Also now you can only merge sub-catchments on adjacent stream branches.

3 ? I have abandoned the TIME noise removal method in the MrVBF module and written my own. Execution time for noise removal has come down from the better part of a day to several minutes. There is now a second pass check box, only use this if you are still getting noise after the first pass.

4 ? After repeated hard to find bugs in calculating the median of the new GRU map that would sometimes cause a crash when committing the MrVBF Raster to the GRU map I have abandoned the TIME method and written my own. It is slightly slower but should be a lot more robust. Execution times will be about 10% longer for this portion.

Note ? The areas displayed in the MrVBF adjust window represent the MrVBF Raster and not the Sub-Catchment map. Therefore they may show larger values.

5 ? In the Generate Scenario Module once a HRU map is dropped if the default fields are present they are automatically selected.

6 ? When running a Scenario much more thorough data checks are done. Look for a full list of error messages in the next version of the documentation.

7 ? The model log now shows the name and full path of all files dropped onto ViewControls

8 ? You can now load data into view controls by right clicking and selecting ?Load? from the context menu.

9 ? I have added 6 new parameters to the results file.
- LateralSaltAlluvial - salt in lateral flow from alluvial
- LateralSaltHillslope - salt in lateral flow from hillslope
- SaltRunOffStorageHillslope ? storage of salt through ET available for runoff on hilllsope
- SaltRunOffStorageAlluvial - storage of salt through ET available for runoff on alluvial
- RunOffSaltAlluvial ? salt in runoff from alluvial
- RunOffSaltHillslope - salt in runoff from hillslope

10 ? I added a catch for bringing in a GRU shape file with out of range values for parameters

11 - Fixed a bug in the results totals for salt3 I was accumulating them instead of averaging

Version 0.9.3

Released: Apr 29, 2005
Licence Date: Mar 16, 2005

What's new in this version?

1 - Added a Windows that pops up after the terrain analysis completes, or after the user merges or deletes polygons. The window shows some statistics of the newly created sub-catchment map such as the size of the largest and smallest sub-catchment and how many sub-catchments if any are smaller in size than the cathment delineation threshold entered by the user. This should alert the modeller of any unwanted small sub-catchments.

2 - Added a new tool to the terrain analysis window. This tool is used at the end of the terrain analysis and zooms the main viewControl the smallest sub-catchment. This also can be used by the modeller to remove any unwanted small sub-catchments.

3 - Added some statistics to the MrVBF window that show the area of the alluvial and hillslope as well as the total area.

4 - From the scenario tab moved the table containing the GRU data to a sepaate tab to make more room.

5 - Changed some of the formatting in the export to CSV I/O layer so that the headers are more readable in Excel without further formatting.

6 - Added a popup form that allows the global changing of all the parameters.

7 - The hillslope store now has a discharge threshold.

8 - Removed the distinction between the unsaturated and aquifer zone thickness. They are now handled by one value in the GRU parameterisation module called depth to basement.

9 - Several more parameters were added to control salt mixing, runoff salt and lateral flow salt. Please see the latest version of the user documentation when it is posted to this site.

All bug reports should be lodged at http://www.toolkit.net.au after first logging on and going to the products page. The latest version of 2CSalt is also available to download from there.

Version 0.9.2

Released: Mar 16, 2005
Licence Date: Mar 16, 2005

What?s new in this release?

1 ? Introduced maximum depth of evaporation from alluvial store. This is set by a new parameter in the GRU table called ?EVAPTHRESH? which is in metres below ground surface.

2 ? Added an alluvial depth discharge threshold. This comes from the gfs map and is now the ?UN_THICK? parameter. So discharge can only occur once the water table gets above the height of the ?AQ_THICK? term.

3 ? Changed the way AQMax and HQMax are calculated. HQMax now uses the perimeter of the alluvial store with a Darcy?s term and AQMax uses twice the length of the major axis of an ellipse fitted to the Alluvial store, also with a Darcy?s term.

4 ? MrVBF has been fixed and should now give similar results to the AML version.

5 ? Bug fix for the volume calculations where the whole GRU area was being used for each store in place of their discreet areas.

6 ? Caught a bug in the ViewControlSelect tool that caused a crash if after running the model from start to finish a GRU was selected from the dropdown list to display a results TimeSeries. This is the bug that caused a crash during the PSC meeting.

7 ? Salt is now a fixed concentration for both stores and is set by the groundwater salinity in the GFS map.

8 ? Added a ?Save Scenario Totals to CSV? button that will export the total Timeseries for each parameter to a row dominant .csv file that can be easily imported into excel.

9 ? Added an option to use TimeResponse theory for the Hillslope aquifer as apposed to the storage discharge relationship. To use this select the ?Use Groundwater Delay? option from the tools menu.

10 ? Changed the way Timeseries are displayed in the Run Scenario tab so that they automatically update after running the model.

11 - SaltExcess2 and SaltExcess3 are now are now at rainfall salinity.

Credit

2CSalt was built on the existing collaborative linkages between CRCCH Industry Partners in Queensland, New South Wales and Victoria and CSIRO in developing and sharing salinity modelling capability, training, skill sharing and information. The activities during the development of 2CSalt were designed to supplement and coordinate existing salt balance modelling within CSIRO and the three Industry Partners.

This development of 2CSalt was cofunded by the Murray-Darling Basin Commission (MDBC) and the CRC for Catchment Hydrology. The financial support of the Murray-Darling Basin Commission as the major funding provider is acknowledged.

The project came under the supervision of a Project Steering Committee established to maintain jurisdictional and Industry Partner involvement under Schedule C of the MDBC Basin Salinity Management Strategy. The following people were part of the Project Steering Committee during the life of the project (Matt Kendall, Bob Newman, Ian Gordon, Stuart Minchin, Dugald Black, Adam Hood, Rory Nathan, Peter Hairsine and Peter Walbrink)

The core project team involved in the development and evaluation of 2CSalt wre (in alphabetical order) Craig Beverly, Brendan Christy, David Freebairn, Mat Gilfedder, Mark Littleboy, Jo Owens, Mark Silburn, Matthew Stenson, Glen Walker and Anna Weeks. During the project the input of Warrick Dawes, Peter Hairsine, Natasha Herron, Narendra Tuteja, Jai Vaze, and Alison Vieritz is recognised.