search

pinskylab / climate_community_crossrealm

Comparisons of community responses to temperature change across realms
MIT License
1 stars 0 forks source link

Verify/standardize mass units from vectraits #4

Closed mpinsky closed 4 years ago

mpinsky commented 4 years ago

extract_VecTraits.R seems to assume that all mass units are in kg, but values in the output (output/mass_BioTrait.csv.gz) seem to be in a mix of g and kg.

For example, these shrimp look likes grams:

                  Species     mass
   1:  Crangon allmanni 1.036453
   2:  Crangon allmanni 1.036453
   3:       Pandalus sp 1.635000
   4: Pandalus borealis 1.635000
   5: Pandalus montagui 1.635000

and these species look like kg:

                        Species         mass
     1:            Pinus strobus 1.875000e+01
     2:     Thalassiosira rotula 1.767146e-09
     3: Thalassiosira eccentrica 1.767146e-09
     4:     Ditylum brightwellii 6.544985e-08
     5: Thalassiosira punctigera 1.767146e-09
mpinsky commented 4 years ago

Hi Malin,

I had a quick look to your issue, and the mass I provided for the species you pointed out to be in kg in the output (Pinus strobus) are actually in grams. So what is in the output file is coherent with what is in the database. However, the more I'm working with this database, the less I trust what is inside. You have information about mass and size and for both you can find a various set of units: "kilogram (wet body mass)", "kilogram (wet tissue mass)" , "kilogram" , "kg", "kilograms" (and also square or cubic meters, etc, etc...)

Which is quite confusing. So far only used what was expressed in a unit related to kilograms. However, I considered that everything is in wet mass (I thought that wet body and tissue mass were the same things, by I know realize that I'm not sure about it...), which might not be the case...

As it seems that you were surprised by some of the values, I don't know if it is trustable. below are some examples of the data I have from vectraits:

If I don't make mistake when I do the conversion (which can happens), it correspond to the values in grams that you thought where in kg (the two last lines are estimated from genus biomass) From that we can think that "kilogram" in the sizei column defines wet body mass, though I'have no guarantee that this is consistent over the entire database.

One important question maybe: did you check all lines to look for strange values or did you picked some randomly? If you made the checks on the entire dataset, maybe we can trust more what is inside.

Also, last point, I noticed a small mistake in my script, it's unlikely that it changes something (maybe on a couple of species) but I'll correct it.

Maybe it can be easier to speak about that by skype or something?

cheers