Difference between revisions of "ARR 2019 Ensembles Processing"
| Line 5: | Line 5: | ||
This page summarizes recommended approach for processing a large amount of ARR 2016 ensembles. | This page summarizes recommended approach for processing a large amount of ARR 2016 ensembles. | ||
==Introduction== | ==Introduction== | ||
| − | The ARR 2016 guidelines implemented ten temporal patterns which are to be applied for each annual exceedance probability (AEP) and duration. This multiplies the number of simulations by ten and requires additional processing in comparison with ARR 1987. The results files can get very large, the creation of the processing tools can lead to tedious work and the actual processing can be time intensive. | + | The ARR 2016 guidelines implemented ten temporal patterns which are to be applied for each annual exceedance probability (AEP) and each storm duration. This multiplies the number of simulations by ten and requires additional processing in comparison with ARR 1987. The results files can get very large, the creation of the processing tools can lead to tedious work and the actual processing can be time intensive. |
==Choice of Output== | ==Choice of Output== | ||
As the model results might get large in size the preferred method to process the ensembles is to use FLT grid output format to export the maximums directly from TUFLOW simulation. FLT files are ESRI binary (float) version of the ASC files and their size is about 1/5 of an ASC file. Binary files can also be processed faster by TUFLOW utilities.<br> | As the model results might get large in size the preferred method to process the ensembles is to use FLT grid output format to export the maximums directly from TUFLOW simulation. FLT files are ESRI binary (float) version of the ASC files and their size is about 1/5 of an ASC file. Binary files can also be processed faster by TUFLOW utilities.<br> | ||
If using only time-series outputs (xmdf, dat) an extra step will be required to export maximum grids (FLT or ASC) with TUFLOW_to_GIS utility. | If using only time-series outputs (xmdf, dat) an extra step will be required to export maximum grids (FLT or ASC) with TUFLOW_to_GIS utility. | ||
| + | ==Median== | ||
| + | Use ASC_to_ASC utility (-statMedian switch) to create median of required values (water level, ...) for each storm duration and each AEP. | ||
| + | ==Maximums== | ||
| + | Use ASC_to_ASC utility (-max switch) to create maximum envelope of the median results for each AEP. | ||
Revision as of 14:44, 11 September 2018
Page Under Construction
This page summarizes recommended approach for processing a large amount of ARR 2016 ensembles.
Introduction
The ARR 2016 guidelines implemented ten temporal patterns which are to be applied for each annual exceedance probability (AEP) and each storm duration. This multiplies the number of simulations by ten and requires additional processing in comparison with ARR 1987. The results files can get very large, the creation of the processing tools can lead to tedious work and the actual processing can be time intensive.
Choice of Output
As the model results might get large in size the preferred method to process the ensembles is to use FLT grid output format to export the maximums directly from TUFLOW simulation. FLT files are ESRI binary (float) version of the ASC files and their size is about 1/5 of an ASC file. Binary files can also be processed faster by TUFLOW utilities.
If using only time-series outputs (xmdf, dat) an extra step will be required to export maximum grids (FLT or ASC) with TUFLOW_to_GIS utility.
Median
Use ASC_to_ASC utility (-statMedian switch) to create median of required values (water level, ...) for each storm duration and each AEP.
Maximums
Use ASC_to_ASC utility (-max switch) to create maximum envelope of the median results for each AEP.