search

courtiol / IsoriX

This is the gthub repository dedicated to the development of the R package IsoriX
12 stars 5 forks source link

Build isoscape over short periods of time #109

Closed PatWright closed 5 years ago

PatWright commented 5 years ago

Hi,

I'm trying to build two isoscapes in Europe based on Spring 2015 and Autumn 2015 data (when samples were collected) from GNIP. After filtering all the data I end up with about 30 points that don't cover my study area and quite weak models. Do you have any suggestions to improve my isoscapes?

Thanks, Patrick

stephkramer commented 5 years ago

Dear Patrick,

the variance in the GNIP data for these two months may be extremely high (not sure where across Europe you sample, but when it is in areas with high precipitation during these months this can happen); also, you might have too few data points per location. I suggest you build an isoscape across many years (and months).

Best wishes Stephanie

Von: PatWright [mailto:notifications@github.com] Gesendet: Mittwoch, 24. Oktober 2018 12:48 An: courtiol/IsoriX Cc: Subscribed Betreff: [courtiol/IsoriX] Build isoscape over short periods of time (#109)

Hi,

I'm trying to build two isoscapes in Europe based on Spring 2015 and Autumn 2015 data (when samples were collected) from GNIP. After filtering all the data I end up with about 30 points that don't cover my study area and quite weak models. Do you have any suggestions to improve my isoscapes?

Thanks, Patrick

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/courtiol/IsoriX/issues/109, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AVUJXUMlTA_Pd0tB0M85RBtO0F5xylmiks5uoEVkgaJpZM4X3pKS.

PatWright commented 5 years ago

Hi Stephanie,

Thanks for your reply. I'll just try more years (2013 to 2017 to start) and that should hopefully solve my problem.

Best wishes, Patrick

courtiol commented 5 years ago

In general, building isoscape using only the data from some months has been shown to be problematic because the ground water does not often just reflect the isotopic signature of present time only but it is influenced by precipitations over several months before... They have been several papers on that which I could dig out if you don't find them. ++