claraqin / neonMicrobe

Processing NEON soil microbe marker gene sequence data into ASV tables.
GNU Lesser General Public License v3.0
9 stars 4 forks source link

Error in downloadRawSoilData() #38

Closed claraqin closed 3 years ago

claraqin commented 3 years ago

I hadn't used downloadRawSoilData() for a few months. Checking on it today, I found that it produces an error when I attempt to download all soil data. While I didn't manage to record the error/warning messages from downloading all data, I did receive the following warning when trying to download DP1.10086 (soil physical properties) for all sites and dates:

Warning message:
In value[[3L]](cond) :
  No data was found for data product DP1.10086.001 at the specified sites and dates.

I used sites="all, startYrMo=NA, and endYrMo=NA.

I plan to investigate this later, but if anyone else has experience with this, please let me know!

claraqin commented 3 years ago

This might be an issue with the underlying data. I downloaded DP1.10086 from the web portal and attempted to stack it using neonUtilities::stackByTable(), but I got the following output and error message:

> neonUtilities::stackByTable("~/Downloads/NEON_soil-periodic.zip")
Unpacking zip files using 1 cores.
  |++++++++++++++++++++++++++++++++++++++++++++++++++| 100% elapsed=02s  
Table(s) sls_soilChemistry bgc_CNiso_externalSummary ntr_externalLab ntr_internalLab ntr_internalLabBlanks ntr_externalSummary are unexpected. Stacking will proceed based on inferred table format; check for updates to neonUtilities.
Stacking operation across a single core.
Stacking table sls_soilCoreCollection
  |++++++++++++++++++++++++++++                      | 56% ~01s          
Error in if (names(df)[i] %in% inVars$fieldName) { : 
  argument is of length zero
lstanish commented 3 years ago

@claraqin Had you tried updating neonUtilities? The functions got an update to accommodate NEON V1 data, that might be causing the problem

claraqin commented 3 years ago

@lstanish At the time, updating neonUtilities did not solve the issue, but I just tried updating it again and it addressed this! Thanks for the reminder.

There's another issue now, though – which is that it only returned soil physical properties, not chemical properties:

> soil_all <- downloadRawSoilData(startYrMo=NA, endYrMo=NA)

...

Returning soil physical variables (but not chemical variables) at the specified sites and dates.
Warning messages:
1: In value[[3L]](cond) :
  No data was found for data product DP1.10078.001 at the specified sites and dates.

Do you have any idea why this might happen?

lstanish commented 3 years ago

@claraqin try the following: test <- loadByProduct(dpID='DP1.10086.001', startdate = '2018-01-01', enddate = '2020-01-01', package='expanded', check.size = FALSE)

claraqin commented 3 years ago

That worked, thanks! This brings up issue #39 (does our code still work with the new NEON soil data product structure?) but I'll close this thread.