Documentation for the Coupled Pacific Ocean Global Atmosphere model (CPOGA)

The CPOGA is a 4.5-layer reduced gravity Pacific Ocean model anomaly-coupled to the NCAR AGCM CAM3. The only modification made to the atmospheric model is a land mask consistent with the ocean model. The ocean model is nested into the atmospheric model by updating SSTs in the Pacific basin with output from the ocean model while all other ocean surfaces are updated using monthly mean climatological SSTs. The climatological SSTs are a composite of the annual cycle for the period 1981-2001, using the global HadISST OI dataset prior to 1981 and the Smith/Reynolds EOF dataset post-1981. A land surface model calculates fields over land (the NCAR CLM2). The atmospheric model is run at T42 resolution with 26 vertical levels (the standard resolution for CAM3). The ocean model extends from 50°S to 60°N and is run at 1 deg X 0.5 deg resolution. The ENSO variability of the updated model is consistent with the documentation provided below.

The ocean model is first run using the CAM3 boundary layer model forced with climatological monthly mean surface air temperature and winds. The main parameters that are varied to get a realistic seasonally-varying climate are ABL, Ocean Mixing, Ocean Diffusion.

Two 150-year CPOGA integrations, with and without air-sea coupling in the extratropical Pacific have been completed. The first simulation is coupled from 55°S to 50°N in the Pacific basin and is referred to as the control integration (CNT). The second simulation is coupled from 15°S to 15°N in the tropical Pacific and is referred to as the tropics-alone integration (TP). In the extratropical Pacific Ocean, TP is forced by heat fluxes and wind stress calculated from climatological surface air temperature and wind (atmospheric anomalies are not passed to the ocean).

The control integration (CNT) compared to Observations

The ocean model consists of four active layers overlying a deep, inert layer of temperature T5=0ºC where the pressure gradient field is assumed to vanish. Each of the layers corresponds primarily to a single water mass type, namely, mixed layer water (T1), upper-thermocline water (T2), lower-thermocline water (T3), upper-intermediate water (T4) in layers-1, -2, -3, and -4, respectively. These layers have thicknesses; H1, H2, H3,and H4. The two figures compare the climatological seasonal means from CNT to LEVITUS94 observations. Observed ocean temperatures in the lower-thermocline are estimated as the average ocean temperatures between H1+H2 and H1+H2+H3.

Four active layers allows for the simulation of both the Tropical Cell, where water recirculates within the tropics in layers-1 and -2, and the Subtropical-Tropical Cell, where water of extratropical origin flows to the equatorial thermocline (layers-2 and -3). Meridional transport in layer-2 shows the equatorward transport of water within 5 degrees of the equator, the lower branch of the Tropical Cell. Meridional transport in layer-3 shows that subtropical water is transported to the tropics from both the North and South Pacific. Note the poleward transport in the eastern subtropical North Pacific in both CNT and observations that blocks the transport of water to the equator through an interior pathway. Observed transports are taken from monthly mean NOAA Pacific Hindcast data from 1980-2004. Observed meridional transports are calculated from meridional velocities integrated over the depth of layers-2 and -3, H2 and H3, respectively.

Mixed-layer physics is parameterized using a Kraus-Turner bulk mixed layer model (Kraus and Turner 1967, Tellus). The adequacy of this parameterization in simulating the mixed layer temperature of a large eddy simulation model was demonstrated by Wang (2003, GRL). CPOGA MAM depth of first model layer, H1, compared to observed LEVITUS94 May mixed layer depth (the depth of the first CPOGA layer is plotted. Model mixed layer depth is deeper in regions where T1 is approximately equal to T2).

CPOGA simulates the observed "re-emergence" of wintertime sea surface temperature anomalies in the North Pacific. This is seen in lead-lag correlations between temperature anomalies in the summertime seasonal thermocline and temperature anomalies between the surface and 150 meters. Correlations are shown for the western North Pacific (38-42°N, 160-180°E), central North Pacific (26-42°N, 164-148°W), and eastern North Pacific 26-42°N, 132-116°W). Reemergence is clearly seen in all three regions with correlations similar to those observed (Figure 12 Alexander et al. 1999, JC). Correlations greater than 0.7 move downward in time, suggesting that temperature anomalies are being subducted into the permanent thermocline.

Transport streamlines in layer-2 (red for transport initiated in the North Pacific between 15-30°N, black for transport initiated in the South Pacific between 30-15°S) show the "potential vorticity barrier" in the North Pacific that blocks transport to the equator through the interior pathway, forcing water the flow around this barrier. To the west of these streamlines, water flows to the western boundary and then to the equator. To the west of these streamlines, water recirculates within the subtropcal gyre and never reaches the tropics. In the South Pacific the flow of water to the equator through interior pathways is unobstructed. Transport streamlines in layer-3 show that water transported from the subtropical South Pacific crosses the equator while water transported from the subtropical North Pacific reaches the Tropics but never flows to the equator in this layer. These streamlines capture the observed differences between pathways in the North and South Pacific, as well as, differences in transport initiated in the eastern and western North Pacific (e.g. Liu and Huang 1998, JPO).

The ocean model includes a mean inflow of 10 Sv (1 Sv = 106 m3s-1) in layer-4 in the southwest corner of the basin and an outflow of 10 Sv divided evenly among layers-1 through layer-3 at the western boundary between 4N-8N to simulate the Indonesian Throughflow. This is required to produce mean transport pathways where intermediate water enters the South Pacific from the Southern Ocean, crosses the equator, and exits the basin in the Indonesian Throughflow, potentially by circuitous paths.

The NINO3 index is calculated by averaging SST anomalies over (150-90°W, 5°S-5°N). CNT has significant power on a broad range of interannual timescales similar to observations. The observed NINO3 index was constructed by the NCAR Climate Analysis Section. A wavelet analysis using 50 years from the CNT integration is plotted to allow for direct comparison with observations.

The spatial structure of ENSO is seen in Regressions of SST anomalies in the tropical Pacific to the CNT NINO3 timeseries. The spatial structure of El Niño in the CNT integration, seen at zero lag in the upper row in Figure 3, is similar to observed with anomalies extending to the eastern boundary and south along the South American coast, as well as, in the off-equatorial South Pacific. Also similar to observations, the La Niña or neutral conditions extend for approximately three years after an El Niño event and have smaller amplitude.

The Tropics-alone run (TP) compared to the control integration (CNT)

Limiting air-sea feedbacks to the tropical Pacific has little impact on the seasonal mean ocean temperatures; mixed layer water (T1), lower-thermocline water (T3), and meridional transports; layer-2  (VH2), layer-3 (VH3).

The impact of excluding extratropical coupling on the frequency of ENSO can be readily seen by comparing the CNT and TP NINO3 time averaged wavelet spectrum. When extratropical coupling is excluded the period of ENSO is essentially locked at 2 years.

Regressions of SST anomalies in the tropical Pacific to the TP NINO3 timeseries demonstrate that the spatial structure of ENSO has changed as well. However, when extratropical coupling is excluded, SST anomalies at the peak of an ENSO event extend farther to the west on the equator, are more meridionally confined in the central equatorial Pacific, are absent along the South American coast, and occur with a predominantly two-year period.

Even though the frequency and spatial structure of ENSO changes when extratropical coupling is excluded, the climate mean SSTs are relatively unchanged. TP vs CNT Sea surface temperatures. The main difference is that wintertime SSTs in the eastern South Pacific are colder when air-sea coupling is allowed in this region. In addition, summertime SSTs are warmer in the western North Pacific when air-sea feedbacks are allowed in this region.