Assessment of hurricane observational under-sampling and its impact on estimated intensity

Periodically issuing estimates of hurricane intensity is an operational requirement for the National Hurricane Center. Because a hurricane's intensity is defined by the maximum near-surface one-minute average wind speed anywhere in the storm, observing this quantity is of paramount importance. However, observing platform limitations, both spatial and temporal, make measuring the peak wind generally difficult, if not impossible. Using a high-resolution (1.3 km) numerical simulation of Hurricane Isabel (2003) – for which it has been shown that 10 m winds from instantaneous model output are nearly equal to one minute average winds at the same locations – a series of simulated observing system experiments are performed to quantify the under-sampling of maximum observed winds relative to the actual (model-produced) peak. We test a number of current aircraft- based observing platforms, including in situ (GPS dropwindsonde, aircraft flight-level data) and remote sensing (SFMR, Doppler radar). Additionally, various flight patterns are examined to optimize observing capability. Finally, intensity estimates using simulated observations from a Hurricane Imaging Radiometer (HIRAD), which is expected to be flown on the NASA GlobalHawk for the 2010 Genesis and Rapid Intensification Process (GRIP) experiment, will be examined in detail.