NOAA National Weather Service provides temporal patterns across the United States.  Temporal patterns are created for 6-hr, 12-hr, 24-hr, and 96-hr durations and expressed in probability terms as cumulative percentages of precipitation totals.  Additionally, the patterns are further divided into 4 quartiles (1st, 2nd, 3rd, and 4th) indicating when the greatest percentage of total precipitation occurs.  A sensitivity analysis was performed using the probability temporal patterns versus historical patterns to review how peak outflows may change depending on the patterns selected.  NOAA Atlas 14 import tutorial is described in detail here: Applying NOAA Atlas 14 Temporal Patterns

Initial Files: SanLorenzo_stormPattern_initial.zip

Excel File: Pattern Comparison Results.xlsx

  

Download NOAA Atlas 14 Temporal Patterns

  1. Head over to the Atlas 14 website: https://hdsc.nws.noaa.gov/pfds/pfds_temporal.html
  2. Click the US map Volume 6.  San Lorenzo falls within Region 7
  3. In the link, under the Volume combo box, select 6: California
  4. Next to the Temporal distribution area select 7
  5. Next to Duration, download the 12-hour, 24-hour, and 96-hour durations. Save the .csv file to your project directory.

Import NOAA Atlas Patterns to HEC-HMS

  1. Import Atlas 14 patterns by opening the SanLorenzo.hms project and heading to Tools | Data | Atlas 14 | Temporal Pattern Importer
  2. In the NOAA Atlas 14 Temporal Pattern Import, select the ca_7_12h_temporal.csv and click Next.  
  3. Select Quartile 1 - Quartile 4 Cases and click Next.  
  4. Save the Paired DSS file to the Project DSS file and choose Restart when the import finishes.
  5. Repeat Steps 1 to 4 with ca_7_24h_temporal.csv and the ca_7_96h_temporal.csv
  6. Navigate to the Watershed Explorer Percentage Curves to see the imported NOAA Atlas 14 patterns

Create Parameter Value Sample

  1. Navigate to Tools | Data | Atlas 14 | Parameter Value Sample Creator 
  2. In the Parameter Value Sample Creator, Specify Storm Pattern in the combo box. Select all 12 hour values (CA_R7_12H*) and click Next.  Name the Parameter Value Sample 12hr_atlas14_all. Save the Paired Data to the Project DSS file and click Restart when complete.  
  3. Repeat Step 2 but with the 24-hr (24hr_atlas14_all) and 96-hr (96hr_atlas14_all) storm pattern durations.
  4. Navigate to this link: https://www.weather.gov/owp/hdsc_publications.  The National Weather Service provides documentation on the development of Atlas 14 precipitation-frequency depths as well as temporal patterns.  Find Vol 6: Precipitation-Frequency Atlas of the United States, California.  Scroll down to Appendix A.6-4 to see a table of precipitation cases for each quartile for each duration

  5. San Lorenzo is in Region 7.  For the 12-hr duration, the Third Quartile has 32% of the cases.  For the 24-hr, the Second Quartile has 30% of the cases.  For the 96-hr, the First Quartile has 36% of the cases.  We will create Parameter Value Samples of the highest percentage cases.
  6. Head back to Parameter Value Sample Creator and repeat Steps 2 but select only the quartile that is most prevalent for the given duration. Name the Parameter Value Samples: 12hr_atlas14_Q3, 24hr_atlas14_Q2, and 96hr_atlas14_Q1.
  7. Once all Parameter Value Samples are created, head back to the Watershed Explorer Parameter Value Samples and review the Parameter Value Samples created.  

Create Uncertainty Analysis Compute

  1. Navigate to the Compute tab and expand the Uncertainty Analysis.  Right click on 1percentAEP-12hr and select Create Copy.  Name the analysis 1percentAEP-12hr_A14_all
  2. Select Parameter 3 and change the Parameter Value to 12hr_atlas14_all
  3. Repeat Steps 1 and 2 for the 24 hour and 96 hour durations.  Make sure to change Parameter 3 to the corresponding duration.  Name the analysis 1percentAEP-24hr_A14_all and 1percentAEP-96hr_A14_all
  4. Right click on 1percentAEP-12hr_A14_all and select Create Copy again.  Name the analysis 1percentAEP-12hr_A14_Q3.
  5. Change Parameter 3 to 12hr_atlas14_Q3.
  6. Repeat Steps 4 and 5 for the 24 hour and 96 hour durations.  Make sure to change Parameter 3 to the corresponding duration.  Name the analysis 1percentAEP-24hr_A14_Q2 and 1percentAEP-96hr_A14_Q1
  7. Start the compute by heading to Compute | Multiple Compute and Select All Uncertainty Analysis simulations.  Click Compute to run the simulation.  

Review Results

  1. In HEC-HMS, in the Results tab, expand 1percentAEP-12hr and the BigTrees node.  Select the Maximum Outflow table.  Under the Statistic Value column copy all 500 values.  
  2. Open the Pattern Comparison Results.xlsx Excel spreadsheet and navigate to the Paste tab.  Paste the 500 values under the 12-hr Historical column.  
  3. Head back to HEC-HMS Results tab and repeat Steps 1 and 2 for 1percentAEP-12hr_A14_all and 1percentAEP-12hr_A14_Q3.  
  4. Repeat Steps 1 and 2 to fill out the rest of the Excel table for the 24-hr and 96-hr patterns
  5. Navigate to Box and Whisker - Histogram tab to view the peak flow outputs.  

Can you use NOAA Atas 14 hypothetical patterns for Design Storm analysis?

When considering only the hydrograph peak value, the 12-hour temporal patterns do not show significant differences in mean, median, or interquartile ranges regardless of whether all Atlas 14 patterns, only Q3 quantiles, or historical storms are used. For shorter duration storms, the choice of pattern type appears to have little impact on results in a Monte Carlo analyses. However, for longer duration events specifically the 24-hour and 96-hour patterns, differences become more apparent. Historical storm patterns produce  higher peak flow values for the mean, median, and interquartile range compared to Atlas 14 patterns. This difference is even greater for the 96-hour temporal patterns. These findings indicate that, for longer storm durations, peak flows can be significantly underestimated if relying solely on NOAA Atlas 14 patterns in Monte Carlo analyses.  Historic storms would be better suited for representing peak flows.

Final Files: SanLorenzo_stormPattern_final.zip

Excel Files: Pattern Comparison Results filled.xlsx