Premium
Evaluation of two cloud parametrization schemes using ARM and Cloud‐Net observations
Author(s) -
Morcrette Cyril J.,
O'Connor Ewan J.,
Petch Jon C.
Publication year - 2011
Publication title -
quarterly journal of the royal meteorological society
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.744
H-Index - 143
eISSN - 1477-870X
pISSN - 0035-9009
DOI - 10.1002/qj.969
Subject(s) - cloud computing , parametrization (atmospheric modeling) , environmental science , cloud fraction , meteorology , climate model , cloud top , cloud height , atmospheric sciences , cloud cover , computer science , climate change , geology , geography , physics , oceanography , quantum mechanics , radiative transfer , operating system
Ground‐based remote‐sensing observations from Atmospheric Radiation Measurement (ARM) and Cloud‐Net sites are used to evaluate the clouds predicted by a weather forecasting and climate model. By evaluating the cloud predictions using separate measures for the errors in frequency of occurrence, amount when present, and timing, we provide a detailed assessment of the model performance, which is relevant to weather and climate time‐scales. Importantly, this methodology will be of great use when attempting to develop a cloud parametrization scheme, as it provides a clearer picture of the current deficiencies in the predicted clouds. Using the Met Office Unified Model, it is shown that when cloud fractions produced by a diagnostic and a prognostic cloud scheme are compared, the prognostic cloud scheme shows improvements to the biases in frequency of occurrence of low, medium and high cloud and to the frequency distributions of cloud amount when cloud is present. The mean cloud profiles are generally improved, although it is shown that in some cases the diagnostic scheme produced misleadingly good mean profiles as a result of compensating errors in frequency of occurrence and amount when present. Some biases remain when using the prognostic scheme, notably the underprediction of mean ice cloud fraction due to the amount when present being too low, and the overprediction of mean liquid cloud fraction due to the frequency of occurrence being too high. Copyright © 2011 Royal Meteorological Society and British Crown Copyright, the Met Office