Extended Abstract (520K)P1.67
The validation of microphysics properties simulated by a cloud resolving model using an in-situ aircraft observation during a cold winter event
Hideaki Ohtake, MRI, Tsukuba, Ibaraki, Japan; and M. Murakami, N. Orikasa, A. Saito, and A. Hashimoto
Japanese Cloud Seeding Experiment for Precipitation Augmentation (JCSEPA), which was the five-year project, was started by the research group composed of the Meteorological Research Institute (MRI) of Japan Meteorological Agency (JMA) and other research institutes since 2006.
The aim of the project is to sophisticate cloud seeding techniques and to establish the method of evaluation of seeding effect on surface precipitation and its impact on the drainage rate of targeted dam catchment. Cloud seeding experiments by the aircraft have been conducted using dry ice pellets in the orographic snow clouds during winter seasons. In the MRI JMA, a research project is now on progress to examine the feasibility of cloud seeding technique enhancing the snowfall in Echigo Mountains which is the main water reservoir for the Tokyo metropolitan area.
To understand the behavior of the cloud physics properties and investigate the model performance with an observational dataset is necessary. Our interest of our present study is to evaluate and validate the hydrometeor properties observed from an aircraft with a numerical simulation by a cloud resolving model (1km-NHM).
In this project, field campaigns on orographic snowfall event had been carried out around the Echigo Mountains in the central part of Japan during winter seasons. We had flown an instrumented aircraft for the consequent 2 winter seasons (on March 2007 and December 2007) over the Echigo Mountains. The instrumented aircraft was equipped with various kinds of cloud microphysics and ordinary meteorological instruments, including three components of wind relative to the ground, on the wing tip pylons and fuselage. The our aircraft was equipped with a Particle Measuring System (PMS) 2-DC(25-800µm) and Forward Scattering Spectrometer Probe (FSSP-100; 2-47µm), the Droplet Measurement Technologies (DMT) the Cloud, Aerosol, and Particle Spectrometer (CAPS ;CAS (0.5-50µm), CIP (25-1550µm)). So far, we made the quality check for the meteorological and microphysics dataset (temperature, pressure, wind speed and direction, cloud water content and cloud particle number concentration etc.). In this field campaign, both in-situ observation and cloud seeding experiment is conducted in one flight.
A cloud seeding model has also been developed based on the Japan Meteorological Agency Non-Hydrostatic Model (JMA-NHM; Saito et al., 2006). The JMA-NHM has five categories of liquid and solid water substances: cloud water, rain, cloud ice, snow, and graupel, as described in Ikawa and Saito (1991). In this numerical simulation, a two-moment bulk parameterization scheme, which prognoses both the mixing ratio and number concentration, is applied to the categories of solid hydrometeor, while one-moment scheme, which prognoses only mixing ratio, is applied to those of liquid hydrometeor. The numerical forecast data of the JMA three dimensional mesoscale model with a 5-km horizontal resolution (5km-NHM) is calculated, and the cloud seeding model with the horizontal resolution of 1 km(1km-NHM) is embedded into the 5km-NHM domain. Compared to the cloud image captured by satellite around the near simulation time, numerical simulations conducted by the forecast group in our institute are also successively performed.
To validate the results of the numerical simulation with an aircraft observation, the observed dataset which is removed the data during the seeding experiment is used. In addition, there is some temporal and spatial difference between the observed and simulated data with a 1km-NHM. Therefore, both observed and simulated data of our all kind of microphysical data (LWC, IWC, its particle concentration and wind fields) are averaged in temporal (1-3h means) and spatial directions because we could compare these dataset at the similar baseline.
Simulated LWC tend to underpredict rather than that of an aircraft observation in all cases (total 21 flights). Although simulated IWC is consistent with that of the observed one, the simulated IWC overpredict by 5 times. Cloud particle concentration of simulated cloud water is consistent with that of the in-situ observation. On the other hand, cloud particle concentration of cloud ice and graupel underpredict too much. In general, simulated concentration of all cases tends to be underpredict.
Horizontal winds speed and its direction are consistent with that of the observed one, and simulated wind fields using a 1km-NHM are reproduced well. On the other hand, observed vertical air motion has a large variation, and averaged vertical air motion is larger than that of the simulated one as same as one order degree. Although we use the non-hydrostatic model with a high horizontal resolution (1km-NHM), simulated vertical wind is not reproduced the actual vertical wind conditions. This means that our model of the present version would have to be improved in the future in this point.
Furthermore, the results of both 5km-NHM and 1km-NHM are compared to investigate the horizontal resolution of our model. Simulated number concentration of cloud ice with 5km-NHM is well reproduced over the Sea of Japan. On the other hand, that of 1km-NHM is very few. Number concentration of cloud ice with 1km-NHM tends to be less than that of 5km-NHM.
Unfortunately, the model setting of the two winter seasons (on March and December 2007) is different in one point in our model simulation. The simulation on Mar 2007 event has no generation process of the ice nuclei in the microphysics process, on the other hand, the Dec 2007 event include that process. Compared the number concentration of the cloud water, LWC and IWC during the two seasons, it is indicated that the generation process of the ice nuclei has a large effect to the simulated results. To conduct recalculation with the generation process of the ice nuclei on March 2007 season is needed and reevaluated.
Poster Session 1, Cloud Physics Poster Session 1
Monday, 28 June 2010, 5:30 PM-8:30 PM, Exhibit Hall
Previous paper Next paper