ensembleMetadata.closestLatLon
Locate the rows of a variable that are closest to a latitude-longitude coordinate
Syntax
Description
Given a variable in the state vector, locates the rows of the variable that are closest to a given latitude-longitude coordinate. Returns the indices of these rows in the overall state vector. If multiple rows are tied as the closest, returns all of the tied rows.
This method proceeds by collecting latitude-longitude coordinates for the variable from the “lat” and “lon” dimensions. The distances between the variable’s coordinates and the input coordinate are then calculated using a haversine function. The haversine function accomodates both -180:180 and 0:360 longitude coordinate systems. Thus, you may use either coordinate system, or even a mix of both.
The method determines latitude-longitude coordinates for the variable using the procedure outlined in the “ensembleMetadata.latlon” method. See the documentation of “ensembleMetadata.latlon” for details. Throws an error if the extracted coordinates are complex valued. Allows NaN and Inf coordinates, so long as at least one extracted coordinate is well-defined. If all extracted coordinates include NaN or Inf values, throws an error.
Indicates that the method should extract latitude-longitude coordinates from the columns of the “site” dimension, rather than the “lat” and “lon” dimensions. Here, the “columns” input is a vector with two elements. The first element indicates the column of the “site” metadata that contains latitude coordinates, and the second element is the column with longitude coordinates. See the documentation of “ensembleMetadata.latlon” for additional details on how coordinates are determined from “site” metadata.
rows = obj.closestLatLon(…, ‘exclude’, variableRows)
Indicate rows of the variable that should be excluded from consideration. Note that variableRows are interpreted relative to the variable, rather than the overall state vector. For example, if variableRows=1, the method will exclude the first row of the variable, regardless of its overall position in the state vector. By contrast, the output rows do indicate positions in the overall state vector.
This syntax is often combined with the “ensembleMetadata.variable” command, which returns metadata for individual variables in the state vector. The outputs of that command can be used to search for excluded values within the context of a single variable.