Improving the precipitation forecast over the Eastern Mediterranean using a smoothed time‐lagged ensemble

• DOI: 10.1002/met.1840
• Publication Date: 2019-08-29T15:55:04Z
• Authors: Pavel Khain; Yoav Levi; Alon Shtivelman; Elyakom Vadislavsky; Evgeny Brainin; Nir Stav
• Keywords: Quantitative precipitation forecast; Model output statistics

Abstract In the absence of convection permitting numerical weather prediction (NWP) ensembles, most recent deterministic NWP precipitation forecast is usually addressed. However, exact intensity, location and timing a not always reliable because chaotic nature complexity formation mechanisms. This study examines way to optimize use forecasts for models. More specifically, it suggests using spatially smoothed time‐lagged ensemble (TLE) obtain more forecasts. A global model (integrated system—IFS) regional (COSMO) over Eastern Mediterranean during period 2016–2018 were used analysis. First, paper defines light, light–moderate moderate intensities 6 hr accumulated (6hAP) investigates corresponding definitions 1 (1hAP). Next, fractional skill score (FSS) are estimate optimal spatial smoothing scale three intensity categories 6hAP 1hAP. The FSS also compare COSMO IFS forecasts, analyse degradation with range. It quantitatively shown that useful smaller larger accumulation time intervals. Finally, TLE formed from successive compared forecast. found that, on average, TLEs have better skills both reason this improvement in illustrated case as an example.