[R] question about redfit from dplR

Mark Hall mh@||89446 @end|ng |rom @tt@net
Sun Jul 7 18:12:05 CEST 2019


Hi All,
I've been trying to use the redfit function in dplR on an unevenly spaced time series of oxygen isotope values and keep running into the following error messages.  x is a series of oxygen isotope values (all negative numbers)y is a series of calibrated or estimated radiocarbon dates ranging from 0 BP to 7000 BP.  
the unevenly spaced time series has over 2000 values.  
The code I've used is as follows:
y<-as.numeric(y)
x<-as.numeric(x)redfit.x<-redfit(x,t=y,tType="time")


and the error message--
Error in redfit(x, t = y, tType = "time") :   REAL() can only be applied to a 'numeric', not a 'integer'In addition: Warning message:In redfit(x, t = y, tType = "time") :  Duplicate times in 't', averaging data
So, then I say, okay, maybe the time type is what is throwing it off--

 redfit.x<-redfit(x,t=y,tType="age")
and once again...
Error in redfit(x, t = y, tType = "age") :   REAL() can only be applied to a 'numeric', not a 'integer'In addition: Warning message:In redfit(x, t = y, tType = "age") : Duplicate ages in 't', averaging data
The oxygen isotope data isn't what is causing an issue per se, since redfit will run with that alone (but since it makes assumptions on the time spacing, that tosses any valid results due to the uneven spacing).
And a sample of my data:
x  [1]  -5.62  -8.93  -6.68  -7.25  -4.93  -7.22  -4.96  -6.30  -4.91  -5.83  -4.00  [12]  -5.50  -5.74  -5.91  -6.12  -5.44  -4.93  -5.20  -4.83  -4.91  -4.74  -4.44  [23]  -4.86  -4.86  -4.70  -5.25  -5.77  -5.23  -7.01  -6.56  -4.70  -5.46  -6.96  [34]  -6.61  -5.27  -4.87  -5.26  -5.79  -4.65  -5.44  -7.07  -5.03  -5.91  -6.08  [45]  -7.76  -6.39  -5.43  -5.45  -5.54  ...
y  [1]    2    5    7    9   13   16   18   20   21   24   25   28   31   37   43   47  [17]   51   54   57   63   68   70   72   75   78   80   83   84   87   89   89   93  [33]   95   97   98   99  101  103  106  106  108  110  110  112  115  116  118  123  [49]  124  124  127  128  130 ...
Any advice would be appreciated.  Also, anyone know of another redfit spectral analysis program other than what is in astrochron?
best, Mark Hall




	[[alternative HTML version deleted]]



More information about the R-help mailing list