daily.station.records {iid.test} | R Documentation |
Test for whether daily station data are independent and identically distributed (iid).
Reference:
Benestad, R.E., 2003: How often can we expect a record-event? Climate Research. 23, 3-13 (pdf)
Benestad, R.E., 2004: Record values, nonstationarity tests and extreme value distributions, Global and Planetary Change, vol 44, p. 11-26
The papers are available in the pdf format from http://regclim.met.no/results_iii_artref.html.
Note: Gaps of missing values will bias the results towards low number of record-events. Likewise, large number of dry days (zero precipitation) will also bias the series since if it only rains 1 out of 100 days, then no new record can be set for days without rain. One solution is to apply the test to the maximum monthly/seasonal rainfall or to exclude the record with many dry days.
daily.station.records(obs,element="precip",subsample=5,tolerance=2,remove.zeroes=FALSE,reverse.plot.reverse=FALSE)
obs |
A daily.station.record (The clim.pact-package). |
element |
Decides which element is analysed ("t2m" or "precip"). |
subsample |
Subsampling to reduce the effects of dependencies. |
tolerance |
How many days with missing value is tolerable before the results are significantly biased. |
remove.zeroes |
Remove records with many dry days (rare and occasional precipitation events tend to lead to undercount). |
reverse.plot.reverse |
see iid.test |
list: 'record.density' and 'record.density.rev' for the reverse analysis.
R.E. Benestad
data(obs) daily.station.records(obs,subsample=30)