For each flood season, the user must define the dates during which a flood event is sampled.  Setting the correct flood season is important when model parameters are a function of the time of year, such as the guide curve in HEC-ResSim.  If the date of the sampled event is important to any of the models used, the distribution of events should be determined by analysis the hydrologic conditions of the basin being studied. 

Event Sampling Flood Seasons

From the Event Sampling tab, a user can identify the flood season. The flood season defines when flood events might occur within a year. The flood season is described by a probability distribution of the peak flood dates using one of five distribution types (Uniform (default), Triangular, Normal, Gamma and Empirical). Users can select the number of flood seasons between one (default) or two as described in the Single Flood Season and Multiple Seasons sections, respectfully.

The flood season can be as simple as a Uniform (default) distribution between a minimum and a maximum date, allowing any date in the range with equal likelihood. The flood season can be a more complex distribution with some dates more likely than other dates. When an analytical distribution is chosen (Uniform, Triangular, Normal or Gamma), the user inputs of that distribution must be entered in the Hydrologic Sampling Editor.

The default values for minimum (start) and maximum (end) dates are 1 January (01Jan) and 31 December (31Dec), respectively, though the user may enter their own values to define the flood season bounds. The user can edit the default values manually (DDMMM); or by clicking . A Select Date dialog box will open that will allow the user to select a day and month. For Normal and Gamma distributions, which do not have explicit upper and lower bounds, user-specified start and end dates may truncate the tails of the distribution. 

Note

When the defined flood season distribution spans 1 January, minimum and maximum dates must be re-defined from the defaults. Another distribution option is to create a user-defined non-parametric empirical (graphical) distribution, which requires a table of cumulative probability versus date entered by the user.

The flood season distribution is used by the hydrologic sampling algorithm in the final step when creating a flood event. A randomly sampled date from the flood season distribution is used as the month and day of the flood event peak flow, as defined by the sampled shape set (refer to Hydrographs to Scale (Shape Sets)). Because different locations within a watershed shape set might have different relative peak dates within a time-window, the location with the extreme relative magnitude (smallest sampled exceedance probability) is the location for which the peak is matched to the sampled date.

Once the user is done with edits on the Event Sampling tab, click Apply. This will save the edits and the Hydrologic Sampling Editor remains open so the user can proceed with the definition of a hydrologic sampling alternative. Alternatively, the user can click OK, which saves the hydrologic sampling alternative and closes the Hydrologic Sampling Editor.

Single Flood Season

The default selection is to enter the flood season distribution for a single flood season. To enter the flood season distribution for one season, select One Season (default) from the Hydrologic Sampling Editor, select the distribution type and define the distribution. Refer to Distributions Used for Flood Seasons section for information regarding each distribution type available in the Hydrologic Sampling Editor.

Multiple Seasons

In some regions, flood events may be generated by different storm-producing mechanisms, often at different times of the year. An example is the Pacific Northwest, which has snowmelt runoff floods in the spring and rainfall-driven floods in the winter. These flood types have different durations with different magnitude-frequency relationships, and occur during different seasons.  The Hydrologic Sampler therefore has the ability to specify two flood sampling seasons, each with a complete description of the probabilistic structure of floods, including season distribution, frequency curves, and event shapes.  

By selecting Two Seasons under the Number of Flood Seasons setting, the user may define a second flood season to be sampled and all following settings must be defined for both seasons.  The user may enter names for both seasons, and these names will be used in other tabs where information is required separately for each season.  The only information common to both seasons is the hydrograph locations.  The user may specify Both Seasons to be sampled in a single event or use Random choice of season to sample an event from either season based on a likelihood of each.  

If Two Seasons are used, and sampling of Both Seasons is selected, the two seasons may overlap.  The minimum time between events is used to ensure that the two sampled floods for a given event are adequately spaced.

Distributions Used for Flood Seasons

Uniform Distribution 

Uniform distribution is the default uncertainty distribution.  Under the uniform distribution, the event is equally likely to occur on any day between two dates. The Uniform distribution includes two inputs - Min (Start) and Max (End). Defaults are 01Jan (Min) and 31Dec (Max). The user can edit these values manually (DDMMM), or by clicking . The Select Date dialog box will open that will allow the user to select a day and month. When the defined flood season distribution spans 1 January, minimum and maximum dates must be redefined. Changes made to the inputs immediately updates the Season Distribution Plot box of the Hydrologic Sampling Editor.

Triangular Distribution 

The Triangular distribution includes three inputs – Min (Start), Mode, and Max (End) dates. Defaults are 01Jan (Min), 01Jul (Mode), and 31Dec (Max). The user can edit these values manually (DDMMM), or by clicking the button. The Select Date dialog box will open that will allow the user to select a day and month. When the defined flood season distribution spans 1 January, minimum and maximum dates must be redefined from the defaults. Changes made to the inputs immediately updates the Season Distribution Plot box of the Hydrologic Sampling Editor.

Normal Distribution 

The Normal distribution includes four inputs – Mean, standard deviation (Std Dev.), Min (Start), and Max (End). Default values are 01 Jul (Mean), 20 (Std Dev.), 01Jan (Min), and 31Dec (Max). The user can edit the Mean, Min, and Max values manually (DDMMM); or by clicking . The Select Date dialog box will open that will allow the user to select a day and month. Std Dev. is entered in days. Changes made to the inputs immediately updates the Season Distribution Plot box of the Hydrologic Sampling Editor.

The Normal distribution does not have explicit upper and lower bounds; therefore, the user-specified start and end dates can optionally be used to truncate the tails of the distribution resulting from the user-defined mean and standard deviation. However, the start and end dates must be redefined when the season spans 1 January; an example is provided for a season starting 1 November and ending 30 April.

Gamma Distribution 

The Gamma distribution includes five inputs – Shape, Scale, Shift, Min (Start), and Max (End). Defaults are 2 (Shape), 20 (Scale), 0 (Shift), 01Jan (Min), and 31Dec (Max). The user can edit the Mean and Max values manually (DDMMM); or by clicking . A Select Date dialog box will open that will allow the user to select a day and month. Shape, Scale, and Shift are entered in days. Changes made to the inputs immediately updates the Season Distribution Plot box of the Hydrologic Sampling Editor.

The Gamma distribution type does not have explicit upper and lower bounds; therefore, the user-specified start and end dates can optionally be used to truncate the season distribution resulting from the user inputs. However, the start and end dates must be redefined when the flood season spans 1 January.

Empirical (graphical) Distribution 

For an Empirical, distribution a default Cumulative Probability function is provided. If necessary, the user can edit the Cumulative Probability Function, the Cumulative Probability column values must span from 0.0 to 1.0. The Date column values should be enter in the DDMM format, or by clicking . A Select Date dialog box will open that will allow the user to select a day and month. Changes made to the user inputs immediately update the Season Distribution Plot.

The Cumulative Probability function can be copied from an outside source (e.g., Microsoft^® Excel or a text editor) and pasted into the table. Right-clicking on the table to display a shortcut menu with various commands for editing tables (refer to Hydrologic Sampling Editor Interface, Modifiable Tables for further details).

Season Distribution Plot

A Season Distribution Plot displays the defined season distribution. The plot has two viewing options: PDF (probability density function) or CDF (cumulative distribution function). Toggling between PDF and CDF, changes the plot in the Season Distribution Plot panel. In addition, to view the season distribution plot in a new window, the user can click Plot, or double-click on the plot displayed in the Season Distribution Plot panel, to open the plot in a new window. 

Stratified Sampling

Stratified sampling in the Hydrologic Sampler changes the distribution of points along the frequency curve sampled. This sampling method can be used to increase emphasis in samples along a certain portion of the frequency curve, reducing the number of total events needed to represent the upper end of the curve.  By default stratified sampling is turned off and any point along the frequency curve is equally likely to be sampled (equivalent to a uniform distribution).  However, when stratified sampling is turned on the probabilities sampled along the frequency curve are weighted by a distribution identified by the user.

To enable stratified sampling and change the settings, click the Edit Stratification Settings button to open the Hydrologic Sampling Stratification Settings dialog. Check the Apply Stratification checkbox to turn on stratified sampling and select a distribution (Gumbel (EV1), Normal, Half-Normal, Exponential, and Uniform) and the location to apply stratified sampling to. The parameters for each distribution predefined and cannot be modified by the user. The Smallest exceedance probability of interest setting is used to set the probability at which the final stratification bin should begin.

Stratified sampling divides the frequency curve into bins based on the distribution selected and smallest exceedance probability of interest.  The number of bins are determined by the number of events within the lifecycle (number of years the analysis period spans). For example a 50-year analysis period will include 50 bins in the stratification. This process ensures that each lifecycle has a sample in every bin. Over the course of generating a set of sampled events in a lifecycle, each bin is sampled in order. The width of each bin is used in re-aggregation and summarization of model results. The probability of each bin is saved to a paired data record in the output variable DSS file for the compute.

Note

The stratified sampling options are not available until at least one primary location has been selected (review Primary Locations). Stratified sampling cannot be used with multiple season sampling (e.g. when Two Seasons) is selected. Also, if the Edit Stratification Settings button does not display a checkmark then the Apply stratification checkbox in the Hydrologic Sampling Stratification Setting dialog box was not checked.