search

NWTlter / NWT_CLM

Workflow scripts for running point simulations of CLM at Niwot Ridge using Tvan forcing data
1 stars 7 forks source link

saddle grid observations #16

Closed wwieder closed 3 years ago

wwieder commented 3 years ago

I'm wondering if the script is parsing out DM vs. WM sites for observed soil moisture and temp? A quick look at these plots seems like the observed lines are identical?

image vs. image

wwieder commented 3 years ago

yeah, something's going on with the temp data being synthesized in the saddle grid image image

wwieder commented 3 years ago

@hhollandmoritz seems like the issue is coming in somewhere here, https://github.com/hhollandmoritz/NWT_CLM/blame/9dcd0d439792526e5fe3f7ae341db864c62617d1/flow.obs.R#L650 but I haven't gone through this carefully enough to understand what's going on.

wwieder commented 3 years ago

OK, now it's looking like the issue is somehow related to the raw data. I uncommented the data around line 550, and each of the vegetation communities look identical...

wwieder commented 3 years ago

I think I found it! I'll fill a new PR but this looks promising! image

wwieder commented 3 years ago

deep soil data from TVAN still don't look good, but surface data are better now.