Arguably such differences could be INCB028050 due to domain edge effects, but these are not observed in CONTROL-E40. The wind intensity over Brazil is somewhat overestimated by CONTROL-E40 run, but the slightly more intense winds along the northern edge of the domain and over the Altiplano differ little from ERA-40. CONTROL-EC4 run shows winds over 3m/s almost throughout the center of the domain.During the Austral winter (JJA, Figure 2(c)), the 850hPa circulation again shows SAH’s influence, albeit a somewhat weaker one. Both simulations overall agree with the ERA-40 reanalysis. This time the northwesterly to westerly circulation over southern region of the domain is reasonably well reproduced. Over Bolivia CONTROL-EC4 exhibits slightly weaker winds, as over the domain’s northeastern sector.

As before, the agreement with AM10 simulations is good and actually closer to ERA-40 than to NCEP. Modeled winds along the Pacific coast again tend to differ from the ERA-40 reanalysis.The modeled flows during equinoxes (autumn, MAM, Figure 2(b), and spring, SON, Figure 2(d)) are in good agreement with the ERA-40 reanalysis, though somewhat weaker winds tend to occur over the domain’s southern third. Other issues noted above also apply to equinoxes.Comparison with AM10 PRECIS simulation runs over a more extended domain for the period 1961�C1990, using NCEP reanalysis and the HadAM3P output as drivers (their Figure 4), shows that current simulation results and reanalysis products are in overall good agreement.

ERA-40 reanalysis shows somewhat stronger winds over Bolivia, Paraguay, and northern Argentina than the NCEP one, and both GCM-driven simulations appear to be closer to ERA-40 reanalysis than to the AM10 NCEP one, at least for summer months.Figure 4CRU fields (left column) and CONTROL-EC4 (center column) seasonal precipitation, given as precipitation rate (mmday?1) for (a) summer (DJF), (b) autumn (MAM), (c) winter (JJA) and (d) spring (SON). The bias, that is, CONTROL-EC4 minus …Inspection of ERA-40 and the simulations for the upper tropospheric level show that they are in reasonably good agreement (Figure 3). During summer months (DJF, Figure 3(a)), Cilengitide ERA-40 reanalysis shows the Bolivian High, centered over 16��S, 67��W approximately, the strong southerly flow over central Brazil, associated with summer deep convection, and the Northeast trough [21, 22]. The westerly flow over the southern part of the domain agrees with observations.