NAME: tpwrspc PURPOSE: wrapper for pwrspc.pro allowing input of a tplot variable name. A Hanning window is applied to the input data, and its power is divided out of the returned spectrum. A straight line is subtracted from the data to reduce spurius power due to sawtooth behavior of a background. UNITS ARE (UNITS)^2 WHERE UNITS ARE THE UNITS OF quantity. freq is in 1/timeunits. THUS THE OUTPUT REPRESENTS THE MEAN SQUARED AMPLITUDE OF THE SIGNAL AT EACH SPECIFIC FREQUENCY. THE TOTAL (SUM) POWER UNDER THE CURVE IS EQUAL TO THE MEAN (OVER TIME) POWER OF THE OSCILLATION IN TIME DOMAIN. CALLING SEQUENCE: CALLING SEQUENCE: tpwrspc, varnames, newname=newname,_extra=_extra INPUT: varname = one tplot variable name OUTPUTS: freq_out=freq_out(optional): output frequency abcissas, in a 1-d array power_out = power_out(optional) : output powers at frequency abcissas, in a 1-d array KEYWORDS: newname = if set,give this name to the new data, the default is to append '_pwrspc' to the input name and pass out the name in the newname variable, Unless /overwrite is set overwrite = if set, write the new data back to the old tplot variable, do not set this with newname noline = if set, no straight line is subtracted nohanning = if set, then no hanning window is applied to the input bin = a binsize for binning of the frequency data, the default is 3 notperhz = if set, the output units are simply the square of the input units err_msg = named variable that contains any error message that might occur NOTES: 1. IF KEYWORD notperhz IS SET, THEN POWER IS IN UNITS^2. If notset power is (as normal) in UNITS^2/Hz. 2. Inputs must be 1-dimensional. For example, if you try to call this on a 3-d vector like fgs data, it will not work. call 'split_vec' first, to split the quantity into its components. 27-mar-2007, jmm, jimm.ssl.berkeley.edu HISTORY: $LastChangedBy: $ $LastChangedDate: $ $LastChangedRevision: $ $URL: $