mescla.uncertainty.montecarlo¶
Monte Carlo uncertainty for mixing ratios.
Analytical propagation assumes local linearity and ignores the simplex constraints.
Monte Carlo does neither: it re-solves the constrained problem for every draw, so the
resulting intervals respect sum(f) = 1 and f >= 0 by construction. That
matters most exactly where it is most needed – for samples near the edge of the
mixing hull, where a fraction is pinned at zero and the sampling distribution is
strongly asymmetric.
Joerin et al. (2002) used this approach to represent temporally variable end-members.
Functions
Percentile intervals on mixing ratios by resampling the inputs. |