Clustering of Observed Diurnal Cycles of Precipitation over the United States for Evaluation of a WRF Multiphysics Regional Climate Ensemble

Journal Article · Wed Nov 01 00:00:00 EDT 2017 · Journal of Climate

DOI:https://doi.org/10.1175/JCLI-D-16-0851.1· OSTI ID:1576699

Mooney, P. A. ^[1]; Broderick, C. ^[2]; Bruyère, C. L. ^[1]; Mulligan, F. J. ^[3]; Prein, A. F. ^[1]

---

1. National Center for Atmospheric Research, Boulder, Colorado
2. Department of Geography, Maynooth University, Kildare, Ireland
3. Department of Experimental Physics, Maynooth University, Kildare, Ireland

The diurnal cycle of precipitation during the summer season over the contiguous United States is examined in eight distinct regions. These were identified using cluster analysis applied to the diurnal cycle characteristics at 2141 rainfall gauges over the 10-yr period 1991–2000. Application of the clustering technique provides a physically meaningful way of identifying regions for comparison of model results with observations. The diurnal cycle for each region is specified in terms of 1) total precipitation, 2) frequency of precipitation occurrence, and 3) intensity of precipitation per occurrence on an hourly basis averaged over the 10-yr period. The amplitude and phase of each element of the diurnal cycle was obtained from harmonic analysis and has been compared with the results of a 24-member multiphysics ensemble of simulations produced by the Weather Research and Forecast (WRF) Model on a region-by-region basis. Three cumulus schemes, two radiation schemes, two microphysics schemes, and two planetary boundary layer schemes were included in the ensemble. Simulations of total precipitation showed reasonable agreement with observations in regions where the diurnal cycle is directly influenced by solar radiation, (e.g., the U.S. Southeast), but they were less successful in regions where other factors influence the diurnal cycle (e.g., the central United States). The diurnal cycle of precipitation frequency and intensity showed substantial biases in the simulations of all eight regions, namely, overestimation of occurrences and underestimation of intensities. Simulations were sensitive to the cumulus and radiation schemes but were largely insensitive to either microphysics or planetary boundary layer schemes.

• Sponsoring Organization: USDOE
• Grant/Contract Number: RFP2010UDW001-4802
• OSTI ID: 1576699
• Journal Information: Journal of Climate, Journal Name: Journal of Climate Vol. 30 Journal Issue: 22; ISSN 0894-8755
• Publisher: American Meteorological Society
• Country of Publication: United States
• Language: English