Central America exhibits a distinct seasonal cycle of rainfall, which is objectively defined as having an onset date and a demise date of the wet season on the first and the last day of the year when its daily rainfall exceeds and falls below the annual mean rainfall climatology. This is defined at the granularity of the rainfall analysis dataset. Additionally, the methodology diagnoses the onset/demise dates of the wet season from an ensemble of 1000 members per season by perturbing the original timeseries to obtain robust probabilistic estimates. We show that both onset and demise date variations have a bearing on the seasonal length and seasonal rainfall anomaly but impact them independently of each other. We demonstrate that a seasonal outlook based solely on the onset date variations has useful prediction skills that portend for real-time monitoring of the onset date of the wet season.

Using a novel but a proven strategy to detect Flash Droughts (FDs), we identified four cases out of 30 (~13%) landfalling Atlantic TCs over the period of 2000-2018 Atlantic hurricane seasons that were associated with FD events in their wake. In all four instances of these FD events, the principal causes were found to be the rapid reduction in precipitation and warming of the surface temperature despite the moderate to weak anomalously wet soil prevalent in the wake of the landfalling TC. These FD cases assume significance because they occurred in southeastern US, which is primarily energy limited for evapotranspiration in the summer. Furthermore, all four cases of FD occurred coincidentally with a more widespread, neighboring drought near the landfall region, which suggests that these FD extend these large-scale droughts from the exacerbated conditions in the wake of the landfalling TC.

In this study we present the results of a Regional Coupled ocean-atmosphere Model (RCM) simulation of the South Asian Summer Monsoon (SASM) forced by global atmospheric and oceanic reanalysis at 20 km grid spacing over a period of 25 years (1986-2010). The RCM shows a more realistic alignment of the simulated rainfall along the orographic features of the domain. Furthermore, the RCM simulates the observed feature of convection over continental SASM region being more vigorous with dominance of mixed warm and cold phase hydrometeors in contrast to the dominance of the warm rain process in the neighboring tropical oceans. Similarly, the upper ocean features of contrasting mixed layer and thermocline depths between the northern and equatorial Indian Ocean are also simulated in the RSM-ROMS. Intra-Seasonal Oscillation (ISO) of the SASM at 10-20 and 20-70 days are simulated in the RSM-ROMS with many of the observed features captured like the latter frequency band being of higher amplitude and the meridional propagation being slower in Bay of Bengal compared to that over Arabian Sea. Additionally, RSM-ROMS shows 12.3 Monsoon Low Pressure Systems (LPS) per season that is comparable to 14.6 per season from observations. Furthermore, the intraseasonal contrasts of LPS between the wet and dry spells of ISO is also reproduced in the RSM-ROMS. The simulation of mesoscale LPS and its intraseasonal variability by the RCM amidst its reasonable fidelity of the SASM climatology clearly demonstrates the value addition of downscaling a global reanalysis at 2.5° (~277-km) to 20-km grid spacing.