Future version of Plankton Identifier

Dear group,

I just finished an intensive year of working with the Zooscan here in Brazil and would like to make a suggestion for the new version of Plankton Identifier. One of the major results one can draw from scannend samples are size distributions (biomass spectra). Calculating those "by hand" is laborious and I have realized that Tanagra has a data binning option. My suggestion is therefore to integrate a function to automatically calculate biomass spectra within PkID. Is that technically feasible? Would you like to have such a function added?

Thank you and all the best for the Zooscanning-community!

Sabine

LAPS wrote:My suggestion is therefore to integrate a function to automatically calculate biomass spectra within PkID. Is that technically feasible? Would you like to have such a function added?

Dear Sabine,

Thank you for your suggestion. Actually, it can already be done, at least partially, with current version of PkID. You can indeed build your own data treatment chain using Tanagra, save this data treatment as tdm file, then import this tdm in PkID. In addition, you can calculate any variable derived from original variables using the "add custom variable" tools. Thus, you can calculate a biomass per object using ESD for instance then use this value in a tdm including both recognition algorithm and spectra construction.
Unfortunately, there is some restrictions to tdm that can be imported in PkID (see here for details) and metadata of the PID files header can not be used yet (to get sample volume if not constant for instance). I will try to improve these points.
The reason why i did not provide such tdm file until now is that there is many possibilities depending on what each user want.

We actually compute biomasses using the ellipsoid volume derived from the MAJOR and MINOR measurements.