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

soil temp data look ok in this file #19

Closed wwieder closed 3 years ago

wwieder commented 3 years ago

maybe you can help see what's going on with soil moisture @hhollandmoritz results from this file look fine, tvan_forcing_data_precip_mods_both_towers_2007-05-11_2020-08-11.txt

image

wwieder commented 3 years ago

These data, however, look crappy, there's no evidence of an annual cycle in the soil temperature data: tvan_West_2007-05-09_19-00-00_to_2020-08-11_00-30-00_flux_P.csv image

hhollandmoritz commented 3 years ago

I figured it out! For the observations, I was pulling the data from sensor "tc10" and "tc30" for 10 and 30 cm below the surface. In both towers (but especially the West tower - where the obs are being pulled from) the tc sensors are quite wonky. The "Tsoil" column in tvan_forcing_data_precip_mods_both_towers_2007-05-11_2020-08-11.txt comes from a different sensor (I think? - it would be good to ask John about this) that is also 10cm below the surface and the data from that sensor look just dandy. The raw data from Tvan call that variable "soil_temp".

I will change the variable for soil temperature from "tc10" to "soil_temp" and it should solve our fell field temp observation issues. BUT we will need to view the 30cm depth readings with a sceptical eye. I don't trust the output of any of the tc sensors at this point.