The boreal summer intraseasonal oscillation (BSISO) is a unique mode of organized tropical convection that varies primarily on subseasonal time scales. Unlike its winter counterpart, the Madden-Julian Oscillation, its role as a source of subseasonal predictability for midlatitude weather extremes has not been extensively investigated. This study provides observational and modeling evidence of BSISO impacts on heat extremes over the Northwestern North America (NNA). The results suggest a robust increase in heat wave frequency and in probability of temperature extremes (by up to ~120%) relative to climatological conditions over most of the NNA when the BSISO is in its western North Pacific phases (phases 6-7) during the boreal summer (June–July). Opposite changes are observed when the BSISO convection is centered over the Bay of Bengal (phases 2-3). The BSISO impacts on heat extremes during phases 6-7 largely originates from the amplitudes and patterns of both the BSISO tropical heating and the strength of the subtropical jet. From a linear wave perspective, our findings indicate that the stronger latent heat release resulting from the convection of the BSISO over the tropical eastern Pacific serves as source of the Rossby waves which propagate poleward, leading to enhanced anticyclonic anomalies and high pressure/subsidence over the NNA. This enhanced anticyclonic circulation is maintained by the convergence of anomalous wave activity fluxes, which is supported by a well-defined meridional waveguide due to a stronger subtropical jet. This anticyclonic circulation in turn results in surface warming via adiabatic processes due to a descending motion. The descending motion provides sunny weather, which enables more solar radiation to reach the land surface unimpeded, causing a more increase in the surface air temperature.

The above results are then used as a basis to evaluate the performance of the 28 models within the Coupled Model Intercomparison Project Phase 6 (CMIP6). We found that CMIP6 models capture the average BSISO propagation, but significantly underestimate the BSISO amplitude. Furthermore, the results indicate that almost half of the CMIP6 models can reasonably reproduce the BSISO teleconnection during phases 6-7, but still encounter difficulties in accurately simulating the corresponding maximum temperature over the NNA. Finally, the robustness of the BSISO teleconnection pattern is also examined using a large ensemble of CESM2 simulations to helps alleviate the constraints imposed by the limited length of the available observed data. The findings of this study carry implications for the source of subseasonal-to-seasonal predictions, which has a potential value for various stakeholders, including urban planners or public health officials.