Lag correlations of sea surface temperature anomalies (SSTAs), sea surface height anomalies (SSHAs), subsurface temperature anomalies, and surface zonal wind anomalies (SZWAs) produced by the Flexible Global Ocean-Atmosphere-Land System model: Grid-point Version 2 (FGOALS-g2) are analyzed and compared with observations. The insignificant, albeit positive, lag correlations between the SSTAs in the southeastern tropical Indian Ocean (STIO) in fall and the SSTAs in the central-eastern Pacific cold tongue in the following summer through fall are found to be not in agreement with the observational analysis. The model, however, does reproduce the significant lag correlations between the SSHAs in the STIO in fall and those in the cold tongue at the one-year time lag in the observations. These, along with the significant lag correlations between the SSTAs in the STIO in fall and the subsurface temperature anomalies in the equatorial Pacific vertical section in the following year, suggest that the Indonesian Throughflow plays an important role in propagating the Indian Ocean anomalies into the equatorial Pacific Ocean. Analyses of the interannual anomalies of the Indonesian Throughflow transport suggest that the FGOALS-g2 climate system simulates, but underestimates, the oceanic channel dynamics between the Indian and Pacific Oceans. FGOALS-g2 is shown to produce lag correlations between the SZWAs over the western equatorial Pacific in fall and the cold tongue SSTAs at the one-year time lag that are too strong to be realistic in comparison with observations. The analyses suggest that the atmospheric bridge over the Indo-Pacific Ocean is overestimated in the FGOALS-g2 coupled climate model.
The seasonal cycle and interannual variability in the tropical oceans simulated by three versions of the Flexible Ocean-Atmosphere-Land System (FGOALS) model (FGOALS-g1.0, FGOALS-g2 and FGOALS-s2), which have participated in phases 3 and 5 of the Coupled Model Intercomparison Project (CMIP3 and CMIP5), are presented in this paper. The seasonal cycle of SST in the tropical Pacific is realistically reproduced by FGOALS-g2 and FGOALS-s2, while it is poorly simulated in FGOALS-g1.0. Three feedback mechanisms responsible for the SST annual cycle in the eastern Pacific are evaluated. The ocean-atmosphere dynamic feedback, which is successfully reproduced by both FGOALS-g2 and FGOALS-s2, plays a key role in determining the SST annual cycle, while the overestimated stratus cloud-SST feedback amplifies the annual cycle in FGOALS-s2. Because of the serious warm bias existing in FGOALS-g1.0, the ocean-atmosphere dynamic feedback is greatly underestimated in FGOALS-g1.0, in which the SST annual cycle is mainly driven by surface solar radiation. FGOALS-g1.0 simulates much stronger ENSO events than observed, whereas FGOALS-g2 and FGOALS-s2 successfully simulate the observed ENSO amplitude and period and positive asymmetry, but with less strength. Further ENSO feedback analyses suggest that surface solar radiation feedback is principally responsible for the overestimated ENSO amplitude in FGOALS-g1.0. Both FGOALS-g1.0 and FGOALS-s2 can simulate two different types of El Nin o events - with maximum SST anomalies in the eastern Pacific (EP) or in the central Pacific (CP) - but FGOALS-g2 is only able to simulate EP El Nino, because the negative cloud shortwave forcing feedback by FGOALS-g2 is much stronger than observed in the central Pacific.
The latest two versions of the IAP Flexible Global Ocean-Atmosphere-Land System (FGOALS) model- versions g1.0 and g1.1, are described in this study. Both two versions are fully coupled GCMs without any flux correction, major changes for g1.1 mainly lie in four aspects: (1) advection schemes for tracer in the ocean component model; (2) zonal filter scheme in high latitudes in the ocean component model; (3) coupling scheme for fresh water flux in high latitudes; and (4) an improved algorithm of airsea turbulent flux depending on the surface current of the ocean. As a result, the substantial cold biases in the tropical Pacific and high latitudes are improved by g1.1, especially g1.1 simulates more reasonable equatorial thermocline, poleward heat transport, zonal overturning stream function in the ocean and sea ice distribution than g1.0. Significant ENSO variability are simulated by both versions, however the ENSO behavior by g1.0 differs from the observed one in many aspects: about twice ENSO amplitude as observed, false ENSO asymmetry, only one peak period around 3 years, etc. Due to improved mean climate state by g1.1, many basic characteristics of ENSO are reproduced by g1.1, e.g., more reasonable ENSO amplitude, two peaks of power spectra for ENSO events, and positive SST skewness in the eastern Pacific as observed.
The Flexible Global Ocean-Atmosphere-Land System Model-gamil (FGOALS-g) was used to study the spring prediction barrier (SPB) in an ensemble system. This coupled model was developed and maintained at the State Key Laboratory of Atmospheric Sciences and Geophysical Fluid Dynamics (LASG). There are two steps in our hindcast experiments. The first is to integrate the coupled model continuously with sea surface temperature (SST) nudging, from 1971 to 2006. The second is to carry out a series of one-year hindcasts without SST nudging, by adopting initial values from the first step on January 1 st , April 1st , July 1st , and October 1st , from 1982 to 2005. We generate 10 ensemble members for a particular start date (1st ) by choosing different atmospheric and land conditions around the hindcast start date (1st through 10th ). To estimate the predicted SST, two methods are used: (1) Anomaly Correlation Coefficient and its rate of decrease; and (2) Talagrand distribution and its standard deviation. Results show that FGOALS-g offers a reliable ensemble system with realistic initial atmospheric and oceanic conditions, and high anomaly correlation (>0.5) within 6 month lead time. Further, the ensemble approach is effective, in that the anomaly correlation of ensemble mean is much higher than that of most individual ensemble members. The SPB exists in the FGOALS-g ensemble system, as shown by anomaly correlation and equal likelihood. Nevertheless, the role of the ensemble mean in reducing the SPB of ENSO prediction is significant. The rate of decrease of the ensemble mean is smaller than the largest deviations by 0.04-0.14. At the same time, the ensemble system "equal likelihood" declines during spring. An ensemble mean helps give a correct prediction direction, departing from largely-deviated ensemble members.
The observed meridional overturning circulation (MOC) and meridional heat transport (MHT) estimated from the Rapid Climate Change/Meridional Circulation and Heat Flux Array (RAPID/MOCHA) at 26.5°N are used to evaluate the volume and heat transport in the eddy-resolving model LASG/IAP Climate system Ocean Model (LICOM). The authors find that the Florida Current transport and upper mid-ocean transport of the model are underestimated against the observations. The simulated variability of MOC and MHT show a high correlation with the observations, exceeding 0.6. Both the simu-lated and observed MOC and MHT show a significant seasonal variability. According to the power spectrum analysis, LICOM can represent the mesoscale eddy characteristic of the MOC similar to the observation. The model shows a high correlation of 0.58 for the internal upper mid-ocean transport (MO) and a density difference between the western and eastern boundaries, as noted in previous studies.
