Dear @nmileva
thank you for raising this issue.
it seems that the tif file you sent does not match with the information provided in the hdr file, indeed. I tried to generate a hdr file matching with the tif, but it did not work.
Right now, the easiest way I found to deal with your problem is to convert your tif image into a ENVI format. However, the hdr file created does not allow to correctly read the original tif file, so I guess there is a question of data encoding or driver which is not correctly identfied.
You can perform the conversion to ENVI file with terra

rasttif <- terra::rast('S2A_T33UUQ_20220627_test_subset.tif')
dir.create('ENVI')
terra::writeRaster(x = rasttif, filename = 'ENVI/S2A_T33UUQ_20220627_test_subset', filetype = 'ENVI')

then, once you get the resulting binary and hdr file, you can update the information required for biodivMapR to run properly:

HDR <- biodivMapR::read_ENVI_header(HDRpath = 'ENVI/S2A_T33UUQ_20220627_test_subset.hdr')
HDR$wavelength <- c(496.6, 560, 664.5, 703.9, 740.2, 782.5, 835.1, 864.8, 1613.7, 2202.4)
HDR$`band names` <- c('B02', 'B03', 'B04', 'B05', 'B06', 'B07', 'B08', 'B8A', 'B11', 'B12')
biodivMapR::write_ENVI_header(HDR = HDR, HDRpath = 'ENVI/S2A_T33UUQ_20220627_test_subset.hdr')

Sorry, this is not very practical. A new version of biodivMapR should be released soon, with major updates so that there will not be any problem to run it with any raster format you want...
I hope this helps.

Best regards,
jb

PS: the initial error may come from the range of values which is not correctly read and does not correspond to standard S2 reflectance values when using the hdr file you used. Then if you updated the mask with radiometric filtering, it will mask everything in your image, and you will get empty pixel values, I think

Thanks! This worked. As you suggested, I needed to adjust also the radiometric filters to the range of reflectance values I have.