Abstract:
The duration and extreme fluctuations of prolonged wet or dry spells associated with intraseasonal variability during extreme monsoon have devastating impacts on agrarian-based economy over Indian subcontinent. This study examines the potential predictability limit of intraseasonal transitions between rainy to non-rainy phases (i.e., active to break phases) or vice versa over central Indian region during extreme monsoon using very high-resolution (0.25° × 0.25°) daily rainfall datasets. The present study reveals that the transitions from both active to break and break to active conditions are more predictable by ~8 days during the weak monsoon (WM) years compared to the strong monsoon (SM) years. Such asymmetric behavior in the limit of predictability could be linked to the distinct differences in the large-scale seasonal mean background instability during SM and WM years. The achievability of such predictability is further evaluated in a state-of-the-art climate model, the climate forecast system (CFSv2). It is demonstrated that the observed asymmetry in predictability limit could be reproducible in the CFSv2 model, irrespective of its spatial resolution. This study provides impetus for useful dynamical prediction of wet/dry spells at extended range during the extreme monsoon years.