1 استادیار، گروه فیزیک فضا، مؤسسۀ ژئوفیزیک دانشگاه تهران، ایران
2 دانشیار، گروه فیزیک فضا، مؤسسۀ ژئوفیزیک دانشگاه تهران، ایران
عنوان مقاله [English]
Lightning is a characteristic of severe weather and often associated with hail and heavy rainfall. It is a natural hazard with potential threat to human life and considerable damages to aviation structures. Therefore, lightning prediction is critical and the real-time lightning detection systems are able to determine the location of cloud-to-ground (CG) lightning strikes accurately.
Generally many indices are used to predict the thunderstorms such as K-Index (KI), Convective Available Potential Energy (CAPE) and Cloud Physics Thunder Parameter (CPTP) that are based on thermodynamic instability parameters. Lightning Potential Index (LPI) is an advanced index for evaluating the potential for lightning activity introduced by Yair et al. (2010) based on the dynamics and microphysics of clouds. According to Yair et al. (2010), LPI is estimated within the charge separation zone of clouds, between 0oC and 20oC, where the non-inductive mechanism involving collisions of ice and graupel particles in the presence of super-cooled water is dominant (Saunders et al., 1991).
In the current study, the meso-scale Weather Research and Forecasting (WRF) model has been used to predict LPI over the northern part of Iran for two case studies of thundercloud event on 9 December 2013 and 25 May 2014. The WRF model is a fully compressible, nonhydrostatic atmospheric model, which uses a terrain-following hydrostatic vertical pressure coordinates (Skamarock et al., 2008). In the present research, WRF version 3.6.1 is used to simulate historical thundercloud event in Iran region.
The model was run at 36 km, 12 km, 4 km and 1.333 km grid spacing. The inner domain is containing Tehran urban area. The Rapid radiative transfer model (Mlawer et al., 1997) with the Dudhia scheme (Dudhia, 1989) was used to simulate the long- and short-wave radiation, respectively. The Monin-Obukhov scheme was used to simulate surface layer fluxes (Janjic, 1996) and the Mellor-Yamada-Janjic turbulent kinetic energy (TKE) scheme was used to simulate boundary layer fluxes (Mellor and Yamada, 1982; Janjic, 1990, 1994). The land surface fluxes were obtained from NOAA model (Chen and Dudhia, 2001, modified by Liu et al., 2006). The Kain-Fritsch scheme was used on the 36 and 12 km grids to parameterize moist convection (Kain and Fritsch, 1993) and the Thompson microphysical scheme was used on the 4 and 1.333 km grids to parameterize microphysical processes. The simulated values of mixing ratios of hydrometers and vertical velocity have been used to calculate the LPI.
Results were evaluated using Cloud-to-ground (CG) lightning flash data from NASA Lightning Imaging Sensor (LIS) and one of the common indices used for forecasting thunderstorms which rely on stability and thermodynamical indices such as K index. Results show that there is a good consistency of both the location of lightning occurrence between the model outputs and LIS data for both understudied cases. Besides, the LPI gives more localized estimation of the location of lightning occurrence compared to the KI. Since K index is not derived from the microphysical fields, it seems to be much less useful for accurate prediction of lightning. Thus LPI provided important information to predict the potential for lightning.