Release Notes: MEDLI Version 2.1

[medlinet]

Bug fixes
• The groundwater nitrate concentration data was not being reported, so the report was showing only zero values. Many thanks to the client who spotted this bug. This has now been fixed.
• The irrigation application option “To specified proportion of requirement to drained upper limit” was assigned the incorrect units.
Impact: Large where the irrigation application choice “Proportion of amount to reach upper storage limit” was chosen, with scenario key outputs incorrectly indicating poor performance.
• A client found an unexpected situation where a fully mulched but bare saturated soil profile would cause the modelled soil water profile to remain locked in that state, preventing any replanting. This has been fixed.
• Another unexpected situation was found where no irrigation would take place when the plant died and irrigation soil water deficit trigger was set to a value below that of the plant available water capacity of the top 100 mm. Since no replanting could take place while the soil remained dry, a rainfall event was the only thing that could end this situation. MEDLI now handles the irrigation of fallow land based on the deficit of evaporable water in the top 150 mm soil depth where irrigation is triggered based on %PAWC or SWD.
Impact: Large where crop death occurs and soil water deficit (or %PAWC trigger equivalent) was set to a value below that of the plant available water capacity of the top 100mm.
• Fix to allow some soil evaporation to take place when there are no plants and the soil is fully mulched (100% total cover). When root depth is zero and total (dead) cover is 100%, evaporation is permitted from 5% of the soil area as if it was bare soil. This prevents the soil water movement from reaching a state of deadlock which prevents irrigation.
• Crash on running a 12-year long scenario now fixed.
• Crash resulting from running a scenario with both no rainfall and no irrigation now fixed
• Finally tracked down and fixed a bug that would cause an occasional “negative uptake rate” error in the plant module.
• Proportion of years above 90% reuse value was being reported in the wrong units. This has been fixed.
• Added a missed limit on a nitrogen mineralisation algorithm to limit mineralisation for soil moisture contents above drained upper limit. The overall impact of this change on predictions was found to be small.
• Fixed inconsistent report header/footer data between the different types of reports
• Climate file validation will now update when a previously missing climate file is imported into the climate data directory
• Fixed various memory leaks and performance issues reducing the overall memory and CPU footprint of the application and speeding up many areas of the application

Settings improvements and user defined libraries
Many changes have been made to the application to allow for enhanced management and customisation of user settings, along with the ability for users to define their own libraries.
• The menu previously only used on the scenario view is now used throughout the application to provide users with a familiar area to access available functionality
• Improved the usability and consistency of many common application commands with improved error handling and general usability (‘New’, ‘Save’ etc.)
• Implemented user defined libraries and improved library validation
• Users can now define their own libraries for use when working with scenarios
• Strict library validation has been implemented for ‘default’ libraries that are included in the application preventing their use in scenarios when modified
• Functionality to allow users to either restore parameters for the current library or save parameters to the current library

Report data serialisation
• The scenario and multirun report have both undergone changes to store the original input scenario as part of the report, this means that users can now extract the exact input file from any given report file, in future release it will also mean that report files can be migrated to the current version.
• The multirun report format has also been updated to vastly improve performance by storing child scenario reports as compressed binary data which can optionally be treated as a separate report or exported as part of the multirun report.

Changes to plant growth module
The plant module has been revised for improved modelling of Vetiver, rotation of plants and nutrient uptake.
• The “Always skip establishment phase of growth cycle” has now been replaced with “Kickstart the establishment phase”. This will kick-start the establishment of Vetiver pasture by resetting the plant biomass and cover to 20% of the biomass and cover following a normal harvest. This accounts for the fact that it is not sown from seed, but is established from slips or splits obtained by subdividing the crown of mature plants.
• Plant mass balance added for plant shoot biomass, nitrogen and phosphorus masses. This allows distinction for the first time between biomass (and plant nutrients) that are actually harvested from the site and that which is lost as residue. Because MEDLI does not model the decomposition of plant residues, the nutrients contained in the residues have always been assumed to be negligible. This may not always be true, particularly in scenarios showing substantial plant stress. The reporting of yield harvested and yield lost as separate values will commence in the next version.
• After harvesting a crop, a crop residue equivalent to 50% of the original soil cover is assumed for soil evaporation purposes rather than assuming no residual cover at all after a crop harvest.
• Some limited uptake of nutrients was found to be taking place when plants were dormant. This has now been further limited to zero since negligible nutrient uptake occurs while plants are dormant.

Impact of all changes to MEDLI modules
The impact of the changes to the plant growth module and the other changes and bug fixes are documented in the MEDLI V2 Version Comparison Reference, accessed from the MEDLI Help View. Briefly:
• For green monthly cover scenarios, V2.1 performs very similarly to V2.0.1.6
• For dynamic crop and pasture scenarios, large impacts are seen. Plant yields generally improve, irrigation generally predicted to be higher in V2.1 than in V2.0.1.6. The impact of these changes was within 10% of the values shown by V2.0.1.6.
• An exception to the above is when scenarios used the irrigation application option “To specified proportion of requirement to drained upper limit”. One Test Scenario (Z8) showed an increase in irrigation from 213 to 1190 mm, with increased yields and reduced pond overtopping.
• Scenarios using Vetiver may also expect to how more realistic in plant yields due to the implementation of the kickstart facility.

Other Changes
• Removed validation preventing climate file data from containing future dates
• Added validation warning to the maximum root depth for pastures and crops when zero is specified, also validation errors and warnings will now also be displayed on the parameter tabs for the paddock parameters.
• Improved the tracking of the last time scenarios and reports were saved such that if a user reverts changes since the last save they will not be warned about unsaved changes
• Improved the behaviour of the undo/redo functionality to show the relevant section of the application when performing undo/redo operations
• Added toaster notifications to display non-critical information to users without requiring any interaction from them
• Implemented key bindings for various commands on scenario and report views that were previously only available using the mouse