2

Possible Duplicate:
Kriging on log transformed rainfall data

I am beginner in R.I had found in literature that prior to perfoming kriging on the data the distribution has to be investigated to check if it is gaussian.

So, inorder to check if the data follows gaussian i plotted the quantile-quantile plot of rainfall for all days using 50 stations data Shown below is the q-q plot for one such day.enter image description here.

The q-q plot of log tarnsformed rainfall is also attached. enter image description here

I have few basic questions:

(1) Though the fit of log-transformed data is good on good rainy days,it is not that on days where rainfall is scanty.

(2) How to back transform the data after interpolation?

(3) How to choose the block size in block kriging? any guidelines. I would like to compare the kriged value against the gridded rainfall from TRMM(25 km by 25 km)?

Thank you

Community
  • 1
user1142937
  • 51
  • 3
  • @RyanThompson I agree, and have flagged it for migration. – Josh O'Brien Jan 11 '13 at 20:40
  • Block kriging is not straightforward for log transformed data see these papers for some methodology [Cressie](http://link.springer.com/article/10.1007%2Fs11004-005-9022-8?LI=true) and [Paul](http://onlinelibrary.wiley.com/doi/10.1111/j.1365-2389.2011.01371.x/full). I have implemented these methods quite easily in `R` in the past, but extremely inefficiently. If I have more time, I will knock something together as an answer. – mnel Jan 11 '13 at 22:28
  • Since the mods don't appear to be moving this, you should feel free to repost it there (perhaps with a little P.S. linking back over here, and a note here linking to there). Thanks for being such a considerate citizen of the community. – Josh O'Brien Jan 12 '13 at 21:04

0 Answers0