Tuesday, 19 April 2016
Plaza Grand Ballroom (The Condado Hilton Plaza)
The representation of the statistical rainfall properties over Darwin is assessed in four versions of the ACCESS forecast model against the dual-polarization CPOL radar observation. Previous results from ACCESS-A show that while the model reproduces the accumulated daily rainfall reasonably well albeit tending to slightly overestimate it, the mean rain rate is too weak, compensated by too large a frequency of occurrence. In terms of intensity the model tends to perform better in the wet than dry regimes, especially during the active monsoon regime, where most of the precipitation occurs. Probability distribution functions of hourly rain rates reveal that while there are very little differences amongst regimes, marked discrepancies are found between model and observations. The model overestimates light rain rates, underestimates moderate to heavy rain rates and produces spurious very heavy rain rates associated with grid point storms. The evaluation of the diurnal cycle of rainfall properties also reveals a general underestimate of rain rates compensated by an overestimated rain frequency over the whole cycle, resulting in overestimated amplitude and poor timing of the diurnal peak. In all regimes, the overestimate of total domain daily rainfall and rain frequency is found to be due to an overestimate of low rain rates produced by the model convective scheme while the underestimate of mean rain rate is due to an underestimate of high rain rates and high rain rates frequency by the model convective scheme. Comparing different horizontal resolutions (40, 12, 4 and 1.5km) of ACCESS, we show that when convection is parameterized, there is a marked land-sea contrast error where there is too much rain over the ocean which abruptly stops when it reaches the coastlines. This land-sea contrast disappears when the convection parameterization is turned off (in the C2 version) allowing the model to develop convection explicitly, but rain peaks are generally too strong.
- Indicates paper has been withdrawn from meeting
- Submission entered in competition