This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Application of NOAA 14 (and proper rainfall distribution type) in Pondpack v8i

Does anybody know how to incorporate NOAA 14 (and its proper rainfall distribution type) in Pondpack v8i? Thanks

Parents
  • Hi nms,

    As you're probably aware, you can obtain cumulative rainfall depth data for a range of events from the NOAA Atlas 14 website (hdsc.nws.noaa.gov/.../pfds). When you look at the data for a specific location, there's also a "Supplemental Information" tab where you can get the dimensionless rainfall distribution data in electronic format.

    Once you have the dimensionless rainfall distribution, you need to add it to the "Dimensionless Rainfall Curves" Engineering Library (Components>Engineering Libraries; right-click Dimensionless Rainfall Curves and create new library). After you create the library, right-click on it and "Add Item." Highlight the new item, and on the right side, click "Dimensionless Depth Collections" and click the Browse button. Follow the prompts to add the dimensionless rainfall distribution. (Note that you can paste data into the table, say from Excel, using CTRL+V). Once added, this dimensionless distribution will be available in the Engineering Library for use with other projects.)

    Once the dimensionless distribution data has been added, you're ready to build the specific storm events. Go to Componenets>Storm Events to open the manager. Add a new item under "Time-Depth." With this new item selected, on the right side of the window, choose the New button and Add Return Event from Dimensionless Curve. Select the Dimensionless curve that you created, and enter additional data (e.g., rainfall depth) as prompted.

    Note that you can also save your Storm Event to an Engineering Library, in case you need it for another project.

    Answer Verified By: Scott Kampa 

  • I have incorporated Kristen's feedback in the below wiki article:

    communities.bentley.com/.../28257.how-to-incorporate-noaa-atlas-14-rainfall-distributions


    Regards,

    Jesse Dringoli
    Technical Support Manager, OpenFlows
    Bentley Communities Site Administrator
    Bentley Systems, Inc.

  • This is a screenshot (sorry for the double monitor) of the csv file I downloaded.  The location is Frostburg, Maryland (18-3415) with a 24 hour temporal duration.  I went to the NOAA site, selected my area, and went to supplemental data as the directions suggested. What I'm getting is similar but in a different format with odd timing.  Not really sure what's going on here.

  • BTW, I looked at Connecticut- the data is the same as in the article.  Maryland's is different.  

  • Thanks for clarifying. It looks like in this case, the distribution is dimensionless on both time and depth and the table is indeed configured differently. So, instead of showing a column of time in hours, there are percentages of total time, in row 13. The percentage numbers (such as "10%") shown in column A are the "percent of occurrence", shown in row 9 of the example in the wiki article.

    The example wiki article looks at the 90% percent of occurrence for the first quartile. To do the same using your data, you would look at row 22. The values shown in the cells in this row correspond to the time percentage in row 13. So for example C22 shows a value of 7.4, and the value for percent of duration above it in cell C13 is 8.3%. 8.3% of 24 hours is 1.992 hours, so the second item in your Rainfall curve time-depth table (after the zero/zero row) would be a time of 1.992 hours and a depth percent of 7.4.

    You could also configure the dimensionless rainfall curve using the "Dimensionless time" option, in which case both the time and the depth columns are entered as a percentage. The time column would be on a scale of zero to 1.0 (representing zero to 100%). So, the 7.4% depth entry in my example above would be used for the dimensionless time of 0.083. When you apply such a rainfall distribution to a storm event later on, you would enter 24 hours as the total time. Since this information seems to be intended for a 24 hour duration though, it might be best to stick with the "Time" option, where the time values are entered in hours (zero to 24).

    I will update the wiki article to explain the steps for both types of rainfall data from NOAA.


    Regards,

    Jesse Dringoli
    Technical Support Manager, OpenFlows
    Bentley Communities Site Administrator
    Bentley Systems, Inc.

  • I managed to get the storms loaded in using the methods described in the wiki.  However, The peak flows are dramatically different compared to the SCS type II distribution we used to use in my locale.  I'm getting a similar hydrograph volume for the rainfall events comparing the NOAA 14 distribution with the SCS distribution, but the peak flows are an order of magnitude different.  I have attached the dimensionless rainfall curve library as well as a copy of the downloaded file from NOAA.  I think the answer is in the time steps in the NOAA file being too large and not capturing  peak accurately.orb_general_24h_temporal md_pa_wv.xlsx

    <NewDataSet><HMIModelingElement><Id>2</Id><LibraryElementType>1</LibraryElementType></HMIModelingElement><DimensionlessRainfallCurve><Id>2</Id><Guid>11f6954e-dbae-4660-9c59-a832ed83c0c2</Guid><DateModified>2017-12-19T17:56:19.5526287-05:00</DateModified><ElementID>0</ElementID><Label>Dimensionless Rainfall Curve -90% first quartile</Label><Notes /><DimensionlessDepthsListCount>13</DimensionlessDepthsListCount><TimeDataType>0</TimeDataType></DimensionlessRainfallCurve><DimensionlessDepthsList><RowID>80</RowID><ParentRowID>2</ParentRowID><DimensionlessDepth>0</DimensionlessDepth></DimensionlessDepthsList><DimensionlessDepthsList><RowID>81</RowID><ParentRowID>2</ParentRowID><DimensionlessDepth>7.4</DimensionlessDepth></DimensionlessDepthsList><DimensionlessDepthsList><RowID>82</RowID><ParentRowID>2</ParentRowID><DimensionlessDepth>21.3</DimensionlessDepth></DimensionlessDepthsList><DimensionlessDepthsList><RowID>83</RowID><ParentRowID>2</ParentRowID><DimensionlessDepth>36.7</DimensionlessDepth></DimensionlessDepthsList><DimensionlessDepthsList><RowID>84</RowID><ParentRowID>2</ParentRowID><DimensionlessDepth>46.2</DimensionlessDepth></DimensionlessDepthsList><DimensionlessDepthsList><RowID>85</RowID><ParentRowID>2</ParentRowID><DimensionlessDepth>51.8</DimensionlessDepth></DimensionlessDepthsList><DimensionlessDepthsList><RowID>86</RowID><ParentRowID>2</ParentRowID><DimensionlessDepth>56.9</DimensionlessDepth></DimensionlessDepthsList><DimensionlessDepthsList><RowID>87</RowID><ParentRowID>2</ParentRowID><DimensionlessDepth>61.6</DimensionlessDepth></DimensionlessDepthsList><DimensionlessDepthsList><RowID>88</RowID><ParentRowID>2</ParentRowID><DimensionlessDepth>66.6</DimensionlessDepth></DimensionlessDepthsList><DimensionlessDepthsList><RowID>89</RowID><ParentRowID>2</ParentRowID><DimensionlessDepth>72.7</DimensionlessDepth></DimensionlessDepthsList><DimensionlessDepthsList><RowID>90</RowID><ParentRowID>2</ParentRowID><DimensionlessDepth>80.4</DimensionlessDepth></DimensionlessDepthsList><DimensionlessDepthsList><RowID>91</RowID><ParentRowID>2</ParentRowID><DimensionlessDepth>89.7</DimensionlessDepth></DimensionlessDepthsList><DimensionlessDepthsList><RowID>92</RowID><ParentRowID>2</ParentRowID><DimensionlessDepth>100</DimensionlessDepth></DimensionlessDepthsList><Time><Id>2</Id><StartTime>0</StartTime><StopTime>24</StopTime><IncrementTime>2</IncrementTime></Time><DimensionlessTime><Id>2</Id><StartDimTime>0</StartDimTime><StopDimTime>100</StopDimTime><DimIncrement>0</DimIncrement></DimensionlessTime><HMIMetaData Guid="bf557eff-6aa2-481f-a1b5-9998c0647965" DateModified="2017-12-19T17:56:19.5526287-05:00" SchemaVersion="5.0.0.1" EngineeringLibraryType="DimensionlessRainfallCurve" /></NewDataSet>

  • Hello Raymond,

    If you add two storms in a single time-depth storm event group, each with the same total depth, but one using your NOAA distribution and the other using SCS Type II 24 hr, you will see the distributions are dramatically different, which is what leads to different runoff peaks despite the same total runoff volume. Here is a comparison using cumulative:

    As you can see, there is a very sharp increase in depth at around the middle of the storm with the SCS Type II, whereas the NOAA event is much more evenly distributed over the 24 hour duration. Here is the resulting runoff hydrograph comparison with one acre catchment with a 90 CN and 10 minute Tc:

    As far as which distribution is best for what application, that will be up to your engineering judgment and design standards. The NOAA distribution you used is definitely very different from the synthetic SCS types, which have a sharp increase in intensity at around the 12 hour mark. Even with the larger time increment with the NOAA distribution, you can still see that it is more evenly distributed over the duration. Perhaps you could look at one of the other quartiles or percent of occurrence.


    Regards,

    Jesse Dringoli
    Technical Support Manager, OpenFlows
    Bentley Communities Site Administrator
    Bentley Systems, Inc.

Reply
  • Hello Raymond,

    If you add two storms in a single time-depth storm event group, each with the same total depth, but one using your NOAA distribution and the other using SCS Type II 24 hr, you will see the distributions are dramatically different, which is what leads to different runoff peaks despite the same total runoff volume. Here is a comparison using cumulative:

    As you can see, there is a very sharp increase in depth at around the middle of the storm with the SCS Type II, whereas the NOAA event is much more evenly distributed over the 24 hour duration. Here is the resulting runoff hydrograph comparison with one acre catchment with a 90 CN and 10 minute Tc:

    As far as which distribution is best for what application, that will be up to your engineering judgment and design standards. The NOAA distribution you used is definitely very different from the synthetic SCS types, which have a sharp increase in intensity at around the 12 hour mark. Even with the larger time increment with the NOAA distribution, you can still see that it is more evenly distributed over the duration. Perhaps you could look at one of the other quartiles or percent of occurrence.


    Regards,

    Jesse Dringoli
    Technical Support Manager, OpenFlows
    Bentley Communities Site Administrator
    Bentley Systems, Inc.

Children
No Data