Detection of weak signals

A professional lightning-detection system should aspire to detect all the lightning strokes which occur within its area of measurement. Only in this manner can a comprehensive prediction be made as to the quality and development of a storm. Intra-cloud strokes, in particular, which are important for the assessment of the strength of a storm, are characterized by low signal strengths. Moreover, even weak cloud-to-ground strokes can cause severe damage.

LINET measures considerably more signals

An ingenious antenna design, highly efficient measurement techniques and filter mechanisms which have been optimized over the years allow for very low thresholds when it comes to the measurement sensitivity of LINET sensors. In this manner, even weak lightning strokes can be detected and located. The significantly higher number of lightning phenomena measured is a tremendous help in the processing of the data, such as Cell Tracking and Nowcasting. The high degree of sensitivity displayed by the sensors also contributes towards the excellent location accuracy of LINET.

Detection of weak lightning strokes in detail

In historical terms, most lightning-detection systems have been specialized to report strong cloud-to-ground strokes. These so-called “return strokes” can usually be recognized thanks to their characteristic field leaps and signal shapes. In this way, algorithms for the discrimination vis-à-vis intra-cloud strokes were also developed. The detection of all lightning strokes as comprehensively as possible, however, also requires the reliable location of weak lightning strokes, which cannot always be clearly distinguished by their signal shape. Hence, nowcast has implemented the location of all lightning signals, independent of the specific signal shape. This has resulted in a significantly higher probability of detection in the case of weak strokes.

Interestingly, there are also cloud-to-ground strokes which occur in their own lightning channel after a classic “return stroke”, and do not display the typical signal shape. These strokes are detected by LINET, but not by the other lightning-detection systems (Stolzenburg et al., “Stroke-Like Upward Illumination Occurring Within a Few Milliseconds After Lightning Return Strokes”, submitted to JGRD, 2012).

The method selected by nowcast is indispensible if intra-cloud strokes are to be efficiently located as well. These strokes namely display manifold signal shapes which cannot be easily categorized. For this reason, LINET offers “Total Lightning”, i.e. the efficient reporting of intra-cloud and cloud-to-ground strokes. This is of substantial significance for the detection of storm developments.

The probability of detection is provided by nowcast using the lightning-current strengths detected (Ampere), however, not in percentage values as is the standard practice otherwise. The reason for this lies in the fact that the total number of lightning strokes actually occurring is not known. With respect to intra-cloud strokes in particular, it is completely unknown how many weak discharges actually take place. However, with the sensor density as recommended by nowcast, a representatively sufficient number of intra-cloud strokes are detected. During the past years, nowcast itself, as well as customers and partners of nowcast, carried out data comparisons with numerous other networks, and always observed that nowcast locates substantially more lightning phenomena than any other low-frequency system.
In 2011, nowcast carried out a data comparison with the special local lightning-detection system of NASA in Florida (Kennedy Space Center), which not only detects lightning strokes in the low-frequency range, but also locates weak lightning discharges (“leader steps”) in the high-frequency range, and hence allows for a highly efficient detection of cloud discharges. It was shown that the nowcast network measures just as many intra-cloud strokes (“cloud flashes”) as the NASA system, and is hence entitled to use the mark of quality “Total Lightning”.

nowcast Partner