Numerical Simulations and Observations of Airflow through the ‘Alenuihāhā Channel, Hawaii

During the summer, sustained winds in the ‘Alenuihāhā Channel, Hawaii, may exceed 20 m s−1 with higher gusts. The Advanced Research Weather Research and Forecasting model is used to diagnose airflow in the Hawaiian coastal waters. High-resolution (2 km) runs are performed for July 2005 covering the ‘Alenuihāhā Channel and nested in a 6-km state domain. Under normal trade wind conditions (7–8 m s−1), winds at the channel entrance are 1–2 m s−1 faster than upstream due to the convergence of the deflected airflows by the islands of Maui and Hawaii, and accelerate through the channel due to along-gap pressure gradients and lower pressure in the wakes of both islands. The acceleration is accompanied by descending airflow (>9 cm s−1) in the exit region with lowering of the trade wind inversion. Deceleration occurs downstream of the channel exit with a rapid change from sinking motion to rising motion (>3 cm s−1). Under normal or strong trade wind conditions, the flow is subcritical [Froude number (Fr) < 1] upstream of the channel, supercritical (Fr > 1) in the exit region, and subcritical again (Fr < 1) downstream with a weak hydraulic jump. The localized sinking motion on the lee side of bordering ridgelines (>1 m s−1) is most significant in the afternoon hours and results in warming and lowering of surface pressure on the lee side, into the channel, and farther downstream. As a result, the channel winds and the wind speed maximum along the southeastern coast of Maui exhibit an afternoon maximum.

Downscaling of Climate Change in the Hawaii Region Using CMIP5 Results: On the Choice of the Forcing Fields*

The Weather Research and Forecasting (WRF) model has been configured as a regional climate model for the Hawaii region (HRCM) to assess the uncertainties associated with the pseudo–global warming (PGW) downscaling method using different warming increments from phase 5 of the Coupled Model Intercomparison Project (CMIP5) model experiments. Results from 15-km downscaling experiments using warming increments from 10 individual CMIP5 models for the two warming scenarios representative concentration pathway 4.5 (RCP4.5) and 8.5 (RCP8.5) are compared with experiments using multimodel mean warming increments. The results show that changes in 2-m temperatures, 10-m wind speed, rainfall, water vapor path, and trade wind inversion vary significantly among the individual model experiments. This translates into large uncertainties when picking one particular CMIP5 model to provide the warming increments for dynamical downscaling in the Hawaii region. The simulations also show that, despite the large interexperiment spread, a single downscaling experiment using multimodel mean warming increments gives very similar results to the ensemble mean of downscaling experiments using warming increments obtained from 10 individual CMIP5 models. Robust changes of the projected climate by the end of the twenty-first century in the Hawaii region shown by most downscaling experiments include increasing 2-m temperatures with stronger warming at higher elevations, a large increase in precipitable water, and an increase in the number of days with a trade wind inversion (TWI). Furthermore, most experiments agree on a reduction in TWI height and an increase in the TWI strength. Uncertainties in the projected changes in rainfall and 10-m wind speed are large and there is little consensus among the individual downscaling experiments.