id summary reporter owner description type status priority milestone component resolution keywords cc
74 Validación de GCMs: Errores de Z crecen con la altura SwenBrands gutierjm "Validando los GCMs de ENSEMBLES contra ERA-40, hemos comprobado que un error dado para Z en 850 hPa crece con la altura. Un revisor del artículo enviado a Climate Research nos da una explicación física de este fenómeno que será de interés para algunos:
""The authors ask me to explain why I understand that the deterioration of performance of geopotential height might be explained on the basis of hydrostatic equilibrium. Starting from the hypsometric equation
P=Ps*exp(-z/H)
we can write:
z=Rd/g*ln(Ps/P) * T
So, errors in T below any level P are being multiplied by the constant Rd/g*ln(Ps/P) before they get converted to errors in z. Thus, as P takes lower values (upper atmospheric levels), this constant grows and errors in Z are higher due to errors in T below that isobaric level (unless errors in temperature cancel in the vertical, something that seems difficult to happen in a systematic way). The value of the constant Rd/g*ln(Ps/P) is almost 11 at 700 hPa and almost 21 at 500 hPa (assuming Ps=1013.25 hPa). I hope this explains what I meant in my previous revision. This is consistent with the result given by the authors in line 290 of the current version of the manuscript.""
" defect new major Otro GCM error chus lluis MarkelGarcia antonio