MetInterp Control File

The MetInterp control file specifies the necessary settings for running MetInterp. Within this control file, each line may have a single argument, with an argument identifier and value separated by a ":" character. Any lines that start with a "*" character, or that are blank, are ignored. Also note that leading or trailing whitespaces are ignored in both the identifier and value.

The settings in the MetInterp file are generally broken up into five groups (although any order of arguments is accepted): General Settings, Gage Settings, Interpolation and Adjustment Settings, Output Settings, Disaggregation Patterns. The sections below describe these argument groups.

General Settings

Gage Settings

Gage settings are broken up into two types: Aggregate Value and DSS Time Series. Aggregate value gages may be used to specify a storm total for a point gage, and a disaggregation pattern. Based on the disaggregation pattern, the specified storm total will be disaggregated to the TimeStep. Alternatively, a DSS time series is converted to the TimeStep (if necessary) with a uniform distribution (or interpolation in the case of instantaneous values).

Aggregate Value Gages

When specifying an Aggregate Value Gage, the identifier is AggregateValueGage. 

Following the identifier, there arguments are in a comma separated list. An example line is below:

AggregateValueGage: true,P1,25,50,2000,1000,10,pattern1

The information in the AggregateValueGage are described below:

DSS Time Series Gages

When specifying a DSS Time Series Gage, the identifier is DssTimeSeriesGage. Following the identifier, there arguments are in a comma separated list. An example line is below:

DssTimeSeriesGage: true,TS1,2000,1000,Gages.dss,/BASIN/LOCATION/PRECIP-INC/01JAN2014/1HOUR/VERSION/

The information in the DssTimeSeriesGage are described below:

MetInterp requires that all DSS Time Series records include coordinate system definitions in their supplemental info. 

Only the Geographic (i.e. lat/lon) Coordinate System is supported by MetInterp.

Missing Value Settings

Interpolation and Adjustment Settings

Interpolation Settings

Interpolation settings are a list of interpolators, each of which has a name, spatial interpolation method, vertical adjustment method, and a temporal sampling method. Any number of interpolators my be specified, but each must have a unique name. An interpolator is defined by a single line, where Interpolator is the identifier. An example line is below:

Interpolator:IDW2,InverseDistanceSquared,None,None

The information in the Interpolator are described below:

Multiple spatial interpolation methods can be computed at the same time by including in the control file one row for the "Interpolator" settings per desired method.

Temporal Sampling Methods are only intended to be used with precipitation, as it is a means to account for the spatial movement of a storm. Although it's only intended to be used with precipitation, MetInterp will not prevent the use of Temporal Sampling with other data.

Bilinear Extrapolation Distance

A special setting for the Bilinear interpolation method is required to specify the extrapolation distance. Other methods use a radius of influence, but the radius of influence concept is incompatible with bilinear interpolation. Since radius of influence is not used, controls over the extrapolation beyond the convex hull of the TIN. This is controlled by an extrapolation distance which is applied orthogonally from the TIN exterior.

Adjustment Settings

Vertical adjustments require settings to be implemented. For bias adjustment, a bias grid is required. For lapse adjustment, an elevation grid and lapse rate must be specified. Interpolation options are specified below:

Output Settings

Output settings specify how the resulting grids will be written. These settings are described in the table below:

Disaggregation Pattern

Disaggregation patterns are only used when Aggregate Value Gages are used. These disaggregation patterns are used to distribute values (typically precipitation depths) over some time period in a non-uniform fashion. For example, a triangular pattern is common, where the majority of precipitation occurs in the middle of the time window. In order to specify these patterns, the "Pattern" identifier is used, followed by a name and array of values, following the example below:

Pattern: Pattern1,{.1, .1, .2, .3, .2, .1}

In the example above, the pattern name is "Pattern1", and the array has 6 values. If the storm is 6 hours in duration, and hourly grids are being created, then the aggregate value for a given gage will be multiplied by each value in the array. For example, if the aggregate value is 5in, the 4th hour will be 1.5in (1.5=5*.3). When the output time steps are not equal to the array length, these 6 values will be converted to the number of time steps necessary. For example, if the storm is 1Day in duration, and the output is hourly, these 6 values will be converted to 24. When converting the number of values, the mass is conserved, meaning that the sum of the resulting 24 values will be equals to the sum of the original 6 values. In the example of 24 hours, the array will be converted to {.025, .025, .025, .025, .025, .025, .025, .025, .05, .05, .05, .05, .075, .075, .075, .075, .05, .05, .05, .05, .025, .025, .025, .025}.