Storm Track: Innovative Tracking of T-storm convective cells.
The thunderstorm is a type of storm produced by a cumulonimbus cloud. It could be characterized by the presence of lightning, heavy rain or snow , hail, strong winds. The basic ingredients used to make a thunderstorm are moisture, unstable air and lift. You need moisture to form clouds and rain. You need unstable air that is relatively warm and can rise rapidly. Finally, you need lift. This can form from fronts, sea breezes or mountains. Many severe phenomena related to them (flash floods, downburst winds) and are very dangerous for air navigation etc..
Knowledge of severe local storms has been increasing rapidly in recent years as a result both of observational studies and numerical modelling experiments. One of the new popular words in meteorologhy has become: nowcasting. Nowcasting comprises the detailed description of the current weather along with forecasts obtained by extrapolation for a period up to 6 hours ahead. In this time range it is possible to forecast small features , such as individual storms with reasonable accuracy. In this contest one of the most used nowcast instruments is the satellite.
Weather satellites are used to get a space-bird's eye view of Earth's atmosphere. Visible and infrrared images are used to detect the amount of water vapour at various altitude or and ground and sea temperatures while winds can be calculated by measuring cloud motion and precipitation amount estimated. Using a combination of the visible and infrared channels, convective clouds and t-storm tracking motions can be detected as a new frontier of the sat monitoring.
Image 1 The pic above represent the three stages of developments of a cumulo nimbus cloud.
Extrapolating data from the satellite we can have a continue and very detailed stream flow info of the size, shape, intensity, speed and direction of movement of individual storms . Thanks to the high resolution of the satellite sensors all the area of operations can be monitored at a resolution of 3 Km updated every 15 minutes. In other words this means having a data on an high resolution basis for virtual point of measurments every 3 kms on a regular grid.
The algorithm works with the sat channels described below:
Use the visible 5,6 and 9 channels
-7.3-10.8>0: cloud base detection (early detection)
-6.2-10.8>0: cloud top detection (Kolios and Feidas)
-Object definition (properties)
The platform allows the detection, monitoring and forecasting of the convection inside a cell , defining the shape, the size and the evolution of the clouds with an intuitive color scheme (yellow triggering, orange growing, blue mature, purple dissipating)
This tool was born to be a satellite data based software but there is also the possibility to associate a grib file with model fields to increase the information (i.e. the height, the pressure of the top or the of the base, the relative humidity, vertical velocity) about the convective cells both in nowcasting and short term forecasting. The wide coverage assured by the Meteosat and the high rate ( 15 minutes ) of sat the images update allows a complete convective clouds step by step evolution monitoring.
Image 1. The tracking system shows the convecttive clouds evoltuton inside active cells on November the 4th 2014 . With a single click are available more useful information like: shape, size, eccentricity and temperature of the convective cell for each grid. An intuitive color shading presentation shows the current and in a next future the forecasted realtime evolution.
Focus on 2014 Genoa flooding events
Frequency of events in northern Italy caused by adverse weather has clearly increased during the past 30 years. Due to heavy rain starting on 9 October 2014 this year several rivers overflowed their basins and a large portion of Genoa,including the center, was flooded.
The severe weather doesn’t bode well for the coming weeks for Italy. The country has suffered some of its most severe flood events at this time of year.
Image 2. The tracking system shows the daily evolution of the convective cells over the Ligurian sea and surroundings on October the 10th 2014.
The above image summarizes the convective cell evolution with steady cells persisting close to Genoa area, evolving and dissipating towards Parma area. Another important cells were indentified along the Rhone Valley. maturating and dissipating close to the Alps Area.
During these severe events (more than 500 mm of rain fallen in few hours) is very difficult for the weather models predict correctlty the rain amount due to the fast convective air masses evolution.
In a very small area the result of severe convective weather storm can be very different with strong winds, hail, heavy rain and not so far away only light rain.
At the same time high resolution weather model not correctly settled-up can exacerbates the convection with false rain amount distribution ( expeccially during summer season ) while global model are not frequently able to detect correctly local convective phenomena due to the weather model resolution.
Radar data ingestion can help to detect native storms and increase locally the accuracy of the prediction on a short time basis but we have to remember that the radar coverage is not so wide while the satellite covers with a regular grid a large portion of the earth.
A good compromise
to evolve the platform would be interface the information of satellite tracking with the other sources (weather stations, lightning data report etc
) to improve the nowcast and the very short term forecasting capability
and accuracy, mitigating at the same time the sat data feed troubles
. Another quite unexplored frontier will be to test the ingestiong and manipulation og these type of data sources into a 3-4DVAR data assimilation system in limited area weather model .
This will be truly an unexplored frontier of the modeling technology to be crucial and innovative way to face the challenges of this era of climate change as shown in the image below, with the sat image taking evidence of the developing of a deep low pressure minina (future TLC) close to Sicily Island.