A STUDY ON THE IMPACTS OF LATENT HEAT PARAMETERIZATION SCHEME ON PREDICTION SKILL OF ENSO WITH A SIMPLE OCEAN-ATMOSPHERE COUPLED MODEL

This study revises Weare＇s latent heat parameterization scheme and conducts an associated theoretic analysis. The revised Weare＇s scheme is found to present potentially better results than Zebiak＇s scheme. The Zebiak-Cane coupled ocean-atmosphere model, initialized by the National Centers for Environmental Prediction and the National Center for Atmospheric Research （NCEP/NCAR） reanalysis of wind stress anomaly at 925 hPa, is referred to as the ZCW coupled model. The atmosphere models of the ZCW coupled model that use Zebiak＇s scheme and the revised Weare＇s scheme are referred to as the ZCW0 and ZCWN atmosphere models, respectively. The coupled ocean-atmosphere models that use Zebiak＇s scheme and the revised Weare＇s scheme are referred to as the ZCWoand ZCWN coupled models, respectively. The simulations between the ZCW0 and ZCWN atmosphere models and between the ZCW0 and ZCWN coupled models are analyzed. The results include： （1） The evolution of heat, meridional wind and divergence anomalies simulated by similar ZCW0 and ZCWN atmosphere models, although the magnitudes of the former are larger than those of the latter; （2） The prediction skill of the Nino3 index from 1982 to 1999 by the ZCWN coupled model shows improvement compared with those by the ZCW0 coupled model; （3） The analysis of E1 Nino events in 1982/1983, 1986/1987, and 1997/1998 and La Nifia events in 1984/1985, 1988/1989, and 1998/2000 suggests that the ZCWN coupled model is better than the ZCW0 coupled model in predicting warm event evolution and cold event generation. The results also show the disadvantage of the ZCWN coupled model for predicting E1 Nino.

Optimization of Selecting Air Conditioning Cold/Heat Sources with Grey Relation Analysis

Selection of air conditioning（AC） cold/heat sources generally concerns about certain aspects and cannot reveal the whole profile of the problems. Grey relation analysis （GRA） is a data processing method to categorize the correlation extent of compared sequences and a certain reference sequence in a system with uncertain information. It is applied to evaluating and selecting AC cold/heat sources from four main aspects, which are technology, economy, reliability, and operation and management. Case study shows that the result for selecting AC cold/heat sources with the GRA method can be more reasonable and convincible. Thus it offers a new approach for designers in heating, ventilating and air conditioning field to compare and evaluate different AC cold/heat sou rces.