Open aappling-usgs opened 7 years ago
do the composite method corrections work with censored values? we should try it.
rich says (and is probably right) that using half the MLD is a bad idea. other options are to draw randomly or otherwise assume a lognormal. this can be a low priority for the time being.
do the composite method corrections work with censored values? we should try it.