Difference between revisions of "QGIS TUFLOW Apply Stability Styling"

From Tuflow
Jump to navigation Jump to search
Line 39: Line 39:
 
[[File: min_max_close_in_time_example.PNG | 450px]]<br>
 
[[File: min_max_close_in_time_example.PNG | 450px]]<br>
 
* Detailed Description Rule 3: The following rule uses a <tt>'''TOLERANCE''</tt> defined as
 
* Detailed Description Rule 3: The following rule uses a <tt>'''TOLERANCE''</tt> defined as
<pre>DIFF_BETWEEN_MAX_AND_TS multiplied by the time series range (time series max - time series min).</pre>
+
<pre>TOLERANCE = DIFF_BETWEEN_MAX_AND_TS multiplied by the time series range (time series max - time series min).</pre>
 
: Flag as unstable if the max value is different than the time series maximum by '''<tt>TOLERANCE</tt>'''. Or flag as unstable if the time of max is different than the time series time of max by '''<tt>TOLERANCE</tt>''' AND the time series value at the time of max is different than the max by '''<tt>TOLERANCE</tt>'''.
 
: Flag as unstable if the max value is different than the time series maximum by '''<tt>TOLERANCE</tt>'''. Or flag as unstable if the time of max is different than the time series time of max by '''<tt>TOLERANCE</tt>''' AND the time series value at the time of max is different than the max by '''<tt>TOLERANCE</tt>'''.

Revision as of 23:55, 12 May 2022

This tool styles the selected layer using a 'traffic light' method to indicate where instabilities could be occurring. This tool currently only supports the _TS.mif (_TS_P.shp) 1D output, however it is intended to expand the scope of this tool in future releases to other inputs and outputs to help with model checking.

Introduced in version 3.3.1'

Running the Tool

The tool can be run by clicking the dropdown arrow next to the 'Apply TUFLOW Style to Current Layer' on the toolbar:
Stability styling toolbar.PNG

Description

The tool uses a 'traffic light' system to indicate where results may be unstable:
Stability styling legend.PNG

The following rules are applied to determine the which category each feature falls within:

_TS_P Output

The _TS_P.shp (_TS.mif) is a 1D timeseries output for Q, V, and H, and includes the following information:

  • Maximum value (from every timestep) and the time of maximum
  • Minimum value (from every timestep) and the time of minimum
  • Time series information - this can be slightly different from the results exported to the 'plot' folder as it will reduce the number of outputs to ensure it conforms with GIS attribute limits
Rules

The rules to categorise 'Likely Unstable' and 'Possibly Unstable' are very similar and the main difference is that they use different sensitivities. The parameters determining how sensitive each is can be modified by the user. The basic rules as follows, a more detailed description and how to change the sensitivity can be found further below:

  • Rule 1: Flag as unstable if the minimum value is less than zero (only applied to "Possibly Unstable")
  • Rule 2: Flag as unstable if the minimum and maximum values occur within a certain timeframe of each other
  • Rule 3: Flag as unstable if the maximum value falls outside of the time series values by a certain amount
  • Rule 4: Flag as unstable if the minimum value falls outside of the time series values by a certain amount
  • Rule 5: Flag as unstable if a number time series 'bounces' with a magnitude greater than a given value exceed a given count



Detailed Description
The following detailed rule description refer to the following variables (which can be changed by the user):

FORWARD_LOOK_COUNT
DIFF_BETWEEN_MAX_AND_TS
UNSTABLE_COUNT_LIMIT
UNSTABLE_PERC_LIMIT


  • Detailed Description Rule 1: Flag as unstable if the minimum value is less than zero (only applied to "Possibly Unstable")
  • Detailed Description Rule 2: Flag as unstable if the minimum and maximum values occur within FORWARD_LOOK_COUNT number of timesteps. This will not be flagged if the minimum value occurs at the first timestep.

Min max close in time example.PNG

  • Detailed Description Rule 3: The following rule uses a 'TOLERANCE defined as
TOLERANCE = DIFF_BETWEEN_MAX_AND_TS multiplied by the time series range (time series max - time series min).
Flag as unstable if the max value is different than the time series maximum by TOLERANCE. Or flag as unstable if the time of max is different than the time series time of max by TOLERANCE AND the time series value at the time of max is different than the max by TOLERANCE.