A previously suggested warm bias effect in the Southern Ocean in moving the zonal mean ITCZ southward is reduced by the cold bias effect at southern latitude. The northern extratropical cold bias appears to be most responsible for southward-shifted zonal mean precipitation, but the zonal mean diagnostic reflects a poor longitudinal pattern of the tropical Pacific response. In (b) the areas are shaded where the response at 99% is not statistically different from zero. confidence level calculated with a two-tailed t-test.
The error bars indicate where the response is not statistically different from zero at the 99 % confidence level using a two-tailed t-test. Hatching in (a,b,d,e) and gray shading in (c,f) indicate where the response is not statistically different from zero at the 99 % confidence level using a two-tailed t-test. Hatching in (d,e,g,h) and gray shading in (f,i) indicate where the response is not statistically different from zero at the 99 % confidence level using a two-tailed t-test.
Hatching indicates where the response is not statistically different from zero at the 99% confidence level using a two-tailed test. In [TRO], the tropical zonal mean surface heat flux is forced and the deviation from the zonal mean is prescribed in TRO*.
Introduction
Motivation
In contrast, the intermodal propagation of the hemispherically antisymmetric bias component of the dual ITCZ largely arises from the intermodal propagation of the tropical asymmetry in the net surface heat flux (Xiang et al., 2017). Meanwhile, the Southeast Pacific (SEP) shows too much precipitation (Bellucci et al., 2010; Oueslati and Bellon, 2015). The SEP precipitation bias has been linked to a model-dependent SST threshold required for the initiation of deep convection (Bellucci et al., 2010).
It has also been attributed to warm sea surface temperature (SST) biases associated with the underestimated stratus cloud fraction off the coast of Peru (Ma et al., 1996), the smoothing of Andean orography in climate models (Takahashi and Battisti, 2007), and weaker than observed cross-country wind (Zheng et al., 2011). In the same context, improved simulation of low-level cloud fractions alleviates the SST and precipitation perturbations in the SEP, but exacerbates the equatorial dry and cold tongues (Fushan et al., 2005), implying a limited improvement in the overall dual-ITCZ bias. Double-ITCZ bias imposes a significant barrier to the simulation of the leading mode of tropical Pacific variability, the El Niño-Southern Oscillation (Guilyardi et al., 2003; Wittenberg et al., 2006) and also to model projections of the Pacific warming pattern under greenhouse gas increases (Seager et al., 2019).
Objectives
Methods
Model and Experiment Design
Stippling indicates the areas where more than two-thirds of 40 CMIP5 models and 52 CMIP6 models exhibit the same sign of SST bias as in CM2.1. We prescribe the q -flux profiles for AM2.1 coupled to a 50-m sheet ocean that has realistic land–ocean distributions and topography. The plate ocean simulation forced by QOBS represents the perfect model climate with no bias (OBS), while that forced by QMODEL represents the model climate with SST bias around the globe (GLO).
While the plate ocean simulations do not perfectly reproduce the corresponding fixed SST simulations, it is worth noting that the SST deviations in GLO and OBS from the corresponding prescribed SST simulations are similar in spatial pattern and magnitude (Figure 2.3), suggesting a common error model. As a result, the SST difference between GLO and OBS (Fig. 2.4) closely matches the actual SST bias (Fig. 2.1a), and the precipitation biases are well reproduced by the plate ocean simulations (contrast Figs. Thus, the plate ocean simulations can be used to relate tropical precipitation biases to surface heat flux bias.
To disentangle the effect of the regional bias of the surface heat flux, we consider a series of q flux profiles, formulated as different combinations of QOBS and QMODEL. For example, the ocean plate simulation with QMODEL between 20°S and 20°N and QOBS over the rest of the globe aims to examine the effect of the tropical surface heat flux bias. As detailed in Table 2.1, we design the experiments to separate the surface heat flux biases in the tropics (TRO), extratropics (EXT), northern extratropics (EXT-N), and southern extratropics (EXT-S) divided at the equator (EXT-SEQ) and poleward (EXT-SPO) of 40°S, as seen in figure 2.2.
We divide the southern extratropics relative to 40°N because the SST bias (Figure 2.1a) and thus the q-flux change sign there (Figure 2.2). In particular, the tropical surface heat flux bias is split into zonally symmetric ([TRO]) and asymmetric (TRO*) components to examine the effect of the zonally averaged diagnostic, which is closely related to the energy framework. This near-linearity of climate responses allows us to adopt regional q -flux simulations to attribute tropical precipitation biases.
We treat the OBS simulation of the ocean plate as an observation, so the difference from the OBS defines the climate bias indicated by the record.
Theory
The so-called equatorial atmospheric energy flux indicates a certain latitude where the zonally averaged and column-integrated atmospheric energy transport (AET) disappears, associated with the ITCZ position (Kang et al., 2008; Donohoe et al. , 2013). This implies that a single ITCZ, associated with an upper-level wind divergence, tends to move meridionally to balance the hemispheric differential energy. For first-order approximations of AET at the equator, as described in Bischoff and Schneider (2016), the location of the ITCZ is expressed as.
Therefore, changes in AET0 at a fixed NEI0 represent meridional shifts of the ITCZ associated with changes in v , and changes in NEI0 at a fixed AET0 lead to ITCZs near the equator or further away associated with changes in v . To illustrate the relationship between dual -ITCZ bias and large-scale energy control, we calculate NEI0 and AET0. Assuming that the energy storage disappears in the annual mean, NEI is calculated as the sum of the downward net radiative energy fluxes at the top of the atmosphere and the upward net energy and heat fluxes at the surface.
AET0 is also calculated in the experiments by integrating the atmospheric NEI over the Southern Hemisphere (SH) and the Northern Hemisphere (NH) and dividing the difference by half.
Results
- Tropical vs extratropical contribution
- Contrasting the effect of zonally symmetric and asymmetric components of tropical bias
- Decomposition of extratropical biases
- Seasonal double-ITCZ problem
The longitude–latitude structure of the tropical precipitation bias reveals rich spatial patterns embedded in the zonal mean (Fig. 2.5b). The tropical surface heat flux bias is responsible for an equatorially elongated cold tongue bias, the north/south tropical Pacific warm/cold bias, a general cold bias in the tropical Atlantic, and a narrow strip of strong warm bias along the western part of the coastlines ( Fig. 3.2a). Noting that the tropical bias manifests itself as a widespread cold bias (Fig. 3.2a), we decompose the tropical QBIAS into the zonal mean and the deviation from the zonal mean.
This zonal asymmetry in the tropical Pacific SST response in [TRO] is due to the contrasting feedback of the shortwave surface SST flux (Lin, 2007) and evaporation damping rate (Xie et al., 2010) over the warm pool and cold tongue. A northern tropical warm bias, partially attributable to the low biased orography over Central America in climate models (Baldwin et al., 2021), causes a strong wet bias over the NI region (Figure 3.3h). This reflects that the high-latitude SH warm bias and the mid-latitude SH cold bias in the CMIP5/6 models (Figure 2.1b) partly originate from a misrepresentation of atmospheric processes.
The GFDL CM2.1 shows the same trend of warm biases poleward of 40°S flanked by cold biases on the equatorward side (Fig. 2.1a), which is also evident in the previous version of the model CM2.0 (Delworth et al. ., 2006). Consistent with the energetic framework, the warm bias poleward of 40°S leads to a negative in EXT-SPO while the cold bias over 20°S-40°S results in a positive in EXT-SEQ (Fig. 3.1 b). Instead, the wet bias appears over the SEP region in EXT-SEQ (Fig. 3.7f) associated with a narrow strip of warm bias extending into the Peruvian coast (Fig. 3.7b) associated with the downward net surface heat flux bias from Chilean side offshore (Fig. 2.2).
However, the NI region presents an opposite precipitation shift associated with the apparent cooling in the northeastern tropical Pacific (Fig. 3.7b). Meanwhile, the Northern Hemisphere shows clear cold biases in the extratropics (Fig. 2.1a), which may be related to a simulated weak AMOC (Wang et al., 2014) and/or cloud biases (Fig. AET0 abnormally northward leads to a Southward shift of tropical rainfall, indicated by a negative (Fig. 3.1b).
The northern extratropical cold bias is advected equatorward, but the cooling response is limited to the north of the equator due to the blocking effect by the mean ITCZ (Fig. 3.7d; Kang et al., 2020). The northern extratropical cold bias is common for CMIP5/6 models (Fig. 2.1b), but it is not emphasized in the context of double-ITCZ bias. A northward precipitation shift over the Indian Ocean in EXT-S, driven by the cold bias over the southern Indian Ocean in EXT-SEQ (Fig. 3.7f), counteracts a southward precipitation shift in other ocean basins.
Concluding remarks
Summary of key points
Discussions
Double ITCZ syndrome in models of coupled general circulation: the role of extensive vertical circulation regimes. Southern Ocean albedo, interhemispheric energy transport, and the dual ITCZ: Global impacts of biases in a coupled model. Climate dynamics. Link between the double intertropical convergence zone problem and cloud bias over the Southern Ocean.
Response of the ITCZ to extratropical thermal forcing: idealized plate–ocean experiments with a GCM. Extratropical-Tropical Interaction Model Intercomparison Project (Etin-Mip): Protocol and initial results. Bulletin of the American Meteorological Society. Tropical biases in the CMIP5 multimodel ensemble: exaggerated equatorial Pacific cold tongue and ITCZ twin problems.
Clouds and Earth's Radiant Energy System (CERES), Energy Balanced and Filled (EBAF), Top-of-Atmosphere (TOA), Release 4.0, Data Product. Journal of Climate. ITCZ double bias in CMIP5 models: interaction between SST, large-scale circulation and precipitation. Double ITCZ bias in CMIP3, CMIP5, and CMI-P6 models based on annual mean precipitation.
Predicting the severity of a spurious “double ITCZ” problem in CMIP5-coupled models from AMIP simulations. 2010), Formation of global warming patterns: sea surface temperature and rainfall. Global precipitation: a 17-year monthly analysis based on measured observations, satellite estimates and numerical model results. Bulletin of the American Meteorological Society. Sea surface temperature biases beneath the stratus cloud deck in the southeastern Pacific Ocean in 19 IPCC AR4-coupled general circulation models.
North Pacific cold temperature biases in CMIP6 simulations and the role of regional vertical mixing.Journal of Climate.
![Figure 2.6 The SST difference of (a) GLO and (b) the sum of the regional q-flux simulations (i.e., [TRO] + TRO* + EXT-N + EXT-SEQ + EXT-SPO) from OBS](https://thumb-ap.123doks.com/thumbv2/123dokinfo/10497389.0/21.892.110.762.130.494/figure-sst-difference-glo-regional-flux-simulations-tro.webp)
![Figure 3.3 The (top) q-flux profile, the difference of (middle) SST and (bottom) precipitation from OBS in (left) [TRO] and (right) TRO *](https://thumb-ap.123doks.com/thumbv2/123dokinfo/10497389.0/28.892.113.777.205.683/figure-flux-profile-difference-middle-sst-precipitation-right.webp)
