search

dmpstats / stochCRM

Graphical User Interface (GUI) developed for a stochastic avian Collision Risk Model (CRM)
7 stars 1 forks source link

flight height data #19

Open MarkTrinder opened 4 years ago

MarkTrinder commented 4 years ago

I think during some of the revisions to the code etc the flight height data may have reverted to old/incorrect versions again.

The random height samples for kittiwake and gannet are the same again. And the gannet heights in the combined sheet (FlightHeight.csv) don't seem to correspond to the height data I have from the BTO work (from here: https://www.bto.org/sites/default/files/u28/downloads/Projects/Final_Report_SOSS02_FlightHeights2014.xls).

bcaneco commented 4 years ago

Hi Mark,

Thanks for spotting (again) the error in the Gannets FHD bootstraps. Revisions in last iteration involved re-issuing the FHD bootstraps from the original Masden files, which contain the erroneous dataset. Our apologies for any inconvenience this may have caused. Online version and GitHub repository have been amended with the correct dataset.

The data in file FlightHeight.csv was also provided to us in the Masden files. That data, however, is not used in any shape in the current stochCRM - only the FHD bootstraps are employed in the simulation. So, as it stands, FlightHeight.csv (and associated code) is just deprecated legacy data. To make it usable, we would have to implement triggers to switch between the two data sources.

Cheers