mitchellmanware
commented
6 months ago @sigmafelix That is odd, and is not by design. I will investigate.
Closed sigmafelix closed 6 months ago
mitchellmanware
commented
6 months ago @sigmafelix That is odd, and is not by design. I will investigate.
mitchellmanware
commented
6 months ago @sigmafelix
process_narr was missing code which subsets available dates (year-month files in path) to dates of interest as defined by date =. The return messaging was unclear because pressure level files have year-month combination but message only sent the working year. Fix incoming soon on branch mm-narr-fix-0321
mitchellmanware
commented
6 months ago @sigmafelix
Patch fixed by filtering available .nc files to unique year-month combinations requested in date =, changing cat() message to include month, and using %Y%m%d format to set terra::time(). See PR https://github.com/NIEHS/amadeus/pull/52
Using same path provided above:
> devtools::load_all()
ℹ Loading amadeus
> d <- amadeus::process_narr(path = "../../../../group/set/Projects/NRT-AP-Model/input/narr/omega", variable = "omega", date = c("2020-03-01", "2020-04-30"))
Cleaning omega data for March, 2020...
Detected pressure levels data...
Cleaning omega data for April, 2020...
Detected pressure levels data...
Returning daily omega data from 2020-03-01 to 2020-04-30.
> d
class : SpatRaster
dimensions : 277, 349, 1769 (nrow, ncol, nlyr)
resolution : 32462.99, 32463 (x, y)
extent : -16231.49, 11313351, -16231.5, 8976020 (xmin, xmax, ymin, ymax)
coord. ref. : +proj=lcc +lat_0=50 +lon_0=-107 +lat_1=50 +lat_2=50 +x_0=5632642.22547 +y_0=4612545.65137 +datum=WGS84 +units=m +no_defs
sources : omega.202003.nc:omega (899 layers)
omega.202004.nc:omega (870 layers)
varnames : omega (Daily Omega on Pressure Levels)
omega (Daily Omega on Pressure Levels)
names : omega~00301, omega~00301, omega~00301, omega~00301, omega~00301, omega~00301, ...
unit : Pascal/s, Pascal/s, Pascal/s, Pascal/s, Pascal/s, Pascal/s, ...
time (days) : 2020-03-01 to 2020-04-30
> head(names(d))
[1] "omega_level=1000_20200301" "omega_level=975_20200301"
[3] "omega_level=950_20200301" "omega_level=925_20200301"
[5] "omega_level=900_20200301" "omega_level=875_20200301" Meant to comment, not close yet 🫣
sigmafelix
commented
6 months ago @mitchellmanware #52 has issues with process_narr tests where I suppose indexing in subset part missed something.
mitchellmanware
commented
6 months ago @sigmafelix Yes just fixed. The filter now detects whether the "time" part of the variable.time.nc file naming convention is YYYY for monolevel or YYYYMM for pressure levels. This should fix the subsetting issue.
mitchellmanware
commented
6 months ago https://github.com/NIEHS/amadeus/pull/52 is good to go. I am sure there are places to clean up the code/make it more efficient but I can deal with this during modularization next week, but good to have a working function. Thank you for bringing this to my attention.
@sigmafelix
process_narrruns at all years regardless of the values ofdateargument.The example below returns different results from what was supposed to be as set in
dateargument:@mitchellmanware Is it as designed or did I miss something in the workflow? More documentation about
dateargument would be helpful.