Earth Networks Lightning Network Performance

The Earth Networks Total Lightning Network (ENTLN) incorporates advanced lightning location technology delivering competitive lightning detection efficiency and location accuracy. ENTLN can detect both in-cloud (IC) and cloud-to-ground (CG) lightning. It consists of over 1600 wideband sensors deployed in 40+ countries to detect lightning and generate faster-localized storm alerts. Since its initial deployment, several improvements were made over the years to enhance its performance and usability. Notable ones are the addition of many new sensors each year to improve detection efficiency and extend lightning detection coverage to any regions of the world. These network expansions have had a dramatic effect on the global detection efficiency of ENTLN. To demonstrate this, we present comparisons with two satellite based optical sensors and contrast these results with similar studies in the past. The two satellite sensors are the Geostationary Lightning Mapper (GLM) and the Lightning Imaging Sensor (LIS). These results also show that ENTLN’s performance has improved compared to a few years ago. The U.S. ENTLN network shows improvements throughout the region. Also, international regions where dense ENTLN networks have been deployed, such as Western and Central Africa as well as Europe, Cambodia, Japan, and Australia, show remarkable improvements. We also compare parameters such as flash peak current and polarity of matches versus non-matches. This analysis is used to better understand regional variations in network performance as well as to diagnose areas where the sensors differ significantly. For most regions, the results are what one might expect, however, there are some unexpected and interesting differences that may reveal inherent biases in ground versus space-based observations.

More recently, in late 2017, Earth Networks began a project to revise the lightning sensor itself, with the goals of improving sensitivity and easing installation requirements. As well as generally updating the hardware, a consolidation of equipment was made, the revised sensor now operates out of a single box, rather than two, and requires a single ethernet cable running to the sensor, significantly reducing installation requirements. In addition, a new firmware was developed to compliment the change in hardware which repackages the data to reduce network traffic, and adds noise rejection logic to improve operation in less than ideal conditions, and provides backwards compatibility with existing sensors. In this study, we show analysis of existing sensors running the new firmware. Initial results show no change in sensor stability, a reduction in network bandwidth by about a factor of two, and a slight increase in sensitivity. The noise reduction logic was found to be effective at sites contaminated with local noise, but to have a slight detrimental effect to sensitivity at sites without noise. As a result, the noise reduction logic is only being deployed to sites with noise contamination issues. We are currently preforming a more in-depth performance analysis of 120 sensors deployed with the new firmware, with plans to update all sensors by the end of 2018.