tindex.rst - pism - [fork] customized build of PISM, the parallel ice sheet model (tillflux branch) (HTM) git clone git://src.adamsgaard.dk/pism (DIR) Log (DIR) Files (DIR) Refs (DIR) LICENSE --- tindex.rst (1201B) --- 1 .. include:: ../../global.txt 2 3 .. _sec-validation: 4 5 Validation case studies 6 ======================= 7 8 9 "Validation" describes the comparison of numerical model output with physical observations 10 in cases where the observations are sufficiently-complete and of sufficient quality so 11 that the performance of the numerical model can be assessed :cite:`Roache`, :cite:`Wesseling`. 12 Roughly speaking, validation can happen when the observations or data are better than the 13 model, so the comparison measures the quality of the numerical model and not merely errors 14 in, or incompleteness of, the data. Because of the difficulty of measuring boundary 15 conditions for real ice flows, this situation is not automatic in glaciology, or even 16 common.\ [#]_ Nonetheless we try two cases, first where PISM is applied on 17 millimeter scale to model a laboratory experiment, and second for a large-scale ice flow 18 in which all uncertainties of bedrock topography, basal sliding, and subglacial hydrology 19 are removed, namely a present-day ice shelf. 20 21 .. toctree:: 22 23 labgum.rst 24 25 ross.rst 26 27 .. rubric:: Footnotes 28 29 .. [#] Which explains the rise of "simplified geometry intercomparisons"; see section 30 :ref:`sec-simp`.