no more srad data from TerraClim?

[urskalbitzer]

Hi,
As usual: thanks a lot for the development and maintenance of this great package!

6-12 months ago, I used the following code without issues.

library(sf)
#> Linking to GEOS 3.11.0, GDAL 3.5.3, PROJ 9.1.0; sf_use_s2() is TRUE
library(climateR)
kan.location <- st_as_sf(data.frame(x = 30.356, y = .562),
                            coords = c("x", "y"), crs = 4326)
srad.kan = getTerraClim(kan.location,
                        param = "srad",
                        startDate = "1970-01-01",
                        endDate = "2020-12-31")

Now, however, I get the following issue:

#> Error in getTerraClim(kan.location, param = "srad", startDate = "1970-01-01", : unused argument (param = "srad")

I tried to use the new argument name varname instead of param but also got an error:

srad.kan = getTerraClim(kan.location,
                        varname = "srad",
                        startDate = "1970-01-01",
                        endDate = "2020-12-31")
#> Error: 'srad' not availiable varname for 'terraclim'. Try: 
#>  > aet [mm] (water_evaporation_amount)
#>  > def [mm] (water_potential_evaporation_amount_minus_water_evaporation_amount)
#>  > PDSI [unitless] (palmer_drought_severity_index)
#>  > pet [mm] (water_potential_evaporation_amount)
#>  > ppt [mm] (precipitation_amount)
#>  > q [mm] (runoff_amount)
#>  > soil [mm] (soil_moisture_content)
#>  > tmax [degC] (air_temperature)
#>  > tmin [degC] (air_temperature)
#>  > vap [kPa] (water_vapor_partial_pressure_in_air)
#>  > vpd [kPa] (vapor_pressure_deficit)
#>  > ws [m/s] (wind_speed)

According to this webpage, "downward surface shortwave radiation" should still be part of that dataset:
https://www.climatologylab.org/terraclimate.html

Do you know what the issue is?

Thanks!

^{Created on 2023-12-08 with reprex v2.0.2}

[mikejohnson51]

Hi @urskalbitzer,

As always thanks for the questions!!... this one is quite annoying... Looking at there server, they chose to use a "-" delimiter rather then a "_" on 2 of their files! srad is one of them... We can get this fixed.

Thanks for the note!

[mikejohnson51]

To add to the fun the OPeNDAP URL is underscore deliminated:

http://thredds.northwestknowledge.net:8080/thredds/dodsC/agg_terraclimate_srad_1958_CurrentYear_GLOBE.nc

[urskalbitzer]

Thanks for the quick response. Glad to hear that the dataset is in principle available and it looks like a relatively easy fix. I will wait for the updated catalog.

[urskalbitzer]

Until this issue is fixed, is there an easy workaround to download the data?

I tried to manually add a respective line to the catalog. However, in this process, the catalog was copied to the global environment and getTerraClim still uses the original one (which is not in the global environment).

Thanks!

[mikejohnson51]

Sorry for the delay - just getting caught up after AGU. Looks like it should now work after a reinstall:

library(sf)
library(climateR)

system.time({
  srad <-
    data.frame(x = 30.356, y = .562) |>
    st_as_sf(coords = c("x", "y"), crs = 4326) |>
    getTerraClim(varname = "srad",
                 startDate = "1970-01-01",
                 endDate = "2020-12-31")
})
#>    user  system elapsed 
#>   0.625   0.077  16.400

plot(srad$date, srad$srad_total, type = "o", pch = 16)
lines(srad$date, predict(lm(srad$srad_total~srad$date)),col='red', lwd = 4)

^{Created on 2023-12-18 by the reprex package (v2.0.1)}

[mikejohnson51]

Hi @urskalbitzer for now I am going to close this but feel free to reopen if its not working for you!

[urskalbitzer]

Hi @mikejohnson51 ,
thanks again for fixing this.
And apologies that my response took a bit longer. I think in the sf- and/or climateR-package some things have changed so that they now interact in a slightly different way (e.g., before I only provided $geometry from a sf-object as the AOI to getTerraClim(), now I have to provide the entire sf object as AOI) . Thus, I had to adapt some additional code to get to the same values as before, but it all worked out well.

[mikejohnson51]

Hey @urskalbitzer! Thanks. Please see 09fd28b. sfc AOI's should be ok again.

Happy holidays!

[urskalbitzer]

Thanks for the quick fix.

Happy holidays to you as well.