Orographic precipitation across an island in southern Norway: model evaluation of time‐step precipitation. (27th December 2012)

• Record Type: Journal Article
• Authors: Barstad, Idar; Caroletti, Giulio N.
• Is Part Of: Quarterly journal of the Royal Meteorological Society. Volume 139:Number 675(2013:Jul.)
• Page Start: 1555
• Page End: 1565
• Publication Date: 2012-12-27
• Subjects: time‐step precipitation -- numerical modelling -- high‐frequency observations -- WRF -- Norway -- Stord; Meteorology -- Periodicals; 551.5
• DOI: 10.1002/qj.2067
• Languages: English
• ISSNs: 0035-9009

Abstract

Abstract: Observations of high spatio‐temporal resolution from a precipitation network across Stord Island, located off the west coast of southern Norway, are compared to state‐of‐the‐art numerical model simulations. The 12 week long observation period shows a clear orographic precipitation signal across the 10–15 km wide island (peak elevation 750 m). The model experiment designed to capture this signal is run with 9–3–1 km nested grid and results are compared with observations at different accumulation intervals. The total amount of precipitation over the 12 week period is underpredicted, even for the 1 km grid. The maximum precipitation intensity, however, is slightly overpredicted. Time‐step (5 s) precipitation from the model is also compared with observed intensities at the highest possible temporal resolution permitted by the rain collection method. The observations indicate that most of the precipitation is formed at intensities from 5 to 20 mm h −1 . A smaller fraction of the precipitation is formed with intensities >20 mm h −1 . The simulated precipitation at the 3 km grid did not reproduce at the correct intensities. The 1 km grid showed an improved tendency to produce the precipitation at the right intensities, but had too high maximum intensities. A test simulation where the intermediate grid had no cumulus parametrization was performed. Even though effects such as undercatchment and unresolved terrain influenced, it was concluded that the test run performed better than the control run. The investigation concluded that, in general, a 1 km grid is sufficient for capturing the most intensive precipitation event in a satisfying way.