The Impact of Supplemental Observations on the Track, Intensity and Structure of Hurricane Isaac in the NCEP Hybrid Gridpoint Statistical Interpolation (GSI) scheme

Previous work examining the impact of synoptic surveillance missions around tropical cyclones (TCs) has focused largely on the improvements to track forecasts in the National Centers for Environmental Prediction (NCEP) Global Forecast System (GFS). However, recent work has shown that these observations can also have an impact on TC structure and intensity when assimilated into the NCEP hybrid ensemble-variational GSI data assimilation scheme, as seen in Tropical Storm Karen (2013).

The Karen study isolated the impact of a single synoptic surveillance mission on the structure, intensity and track of that cyclone. However, additional cases need to be examined to better understand the possible range of impacts of these observations in the new hybrid scheme. Isaac (2012) provides an opportunity to examine the cumulative impact of six cycles of synoptic surveillance data on a long-track tropical cyclone that eventually made landfall in Louisiana as a hurricane.

Preliminary results indicate that there was large cycle-to-cycle variability in the impact of the supplemental observations on the track of Isaac, with some GFS forecast cycles showing large improvements due to the dropwindsonde observations and other cycles showing little impact. On average, the dropwindsondes reduced day 4 and 5 GFS track forecast error by about 30%, with little change seen at earlier forecast lead times. This result contrasts with previous studies that showed the largest impact from the supplemental dropwindsonde data on track during the first 48 hours of the forecast period. Average GFS intensity errors were similar with and without the dropwindsonde data through day 4, with a 25% improvement seen at day 5 with the supplemental observations.