Fasulloand Webster (2003) used a normalized spatial mean of VIMT over the Arabian Seaas Hydrological Onset and Withdrawal Index (HOWI) to identify the Indian summermonsoon onset and withdraw. While the newly developed HOWI index shares abouthalf of its variability with existing onset identifications, markedly increasedand statistically significant relationships emerge between the monsoontransitions, JJAS rainfall, and ENSO when the new index is considered. The HOWIis robust to bogus monsoon onsets and reflects the timing rather than thespatial character of the transitions. Their study also shows that there areparticularly rapid VIMT fluctuations during monsoon onset and withdrawal over theArabian Sea. Taniguchi and Koike (2006) found that the criteria for Indiansummer monsoon onset are more suitable with wind speed as it is easier, clearerand successfully screens out ?Bogus onset.
Sabeerali et al. (2011) studied theonset, withdrawal and Length of the Rainy Season (LRS) of the Indian summermonsoon using Atmospheric General Circulation Model (AGCM) ECMWF Hamburg Modelversion 5 (ECHAM5) for the period 1950 to 2008. Their study reveals that thereis an early withdrawal trend after 1966/67, as a result, the LRS shortened.