I am using the landsat BSL (Bare Soil Line) package from R https://rdrr.io/cran/landsat/man/BSL.html, and I am wondering why the code sample is using llimit 0.005 and not 0.01:

BSL(band3, band4, method = "quantile", ulimit = 0.99, llimit = 0.005)

Source code is at https://rdrr.io/cran/landsat/src/R/BSL.R

I looked at the code and documentation, and documentation says: "The quantile method takes the lowest set of points, those with a NIR/red ratio less than the llimit-th quantile." Is there a reason why llimit parameter is set to 0.005 and not 0.01? Does 0.005 set up a 2 tail region, one at 0.005 and one at 0.99?

I emailed the author of this package but I didn't get a response about this question so I'm hoping someone on GIS SE might know.

  • A 0.5% percentile is just the default, if it does not track with your knowledge or literature that is outlining your research, just change it to the lower limit argument to 1% percentile, which is symmetrical with 99%. BTW, this function is from the landsat package. The BSL package is for Bayesian Synthetic Likelihood with Graphical Lasso. – Jeffrey Evans Nov 27 '18 at 21:59
  • Thanks for clarifying @Jeffrey Evans. Yes, I meant the landsat BSL package. – J. C. Dec 2 '18 at 7:07

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Browse other questions tagged or ask your own question.