The future of contrail verification: Integrating ground observations with ADS-B and weather models

In 2019, the warming effects of contrails – thin clouds formed by aircraft – may have been over twice as large as the warming caused by aviation’s cumulative CO2 emissions since the 1940s.

Scientists use models to estimate the climate impact of contrails and explore ways to reduce their warming effects. However, these models need to be validated with real-world data from sources like satellites and ground-based cameras. While satellites can observe contrails over the large areas, they can struggle to detect very young or faint contrails and older contrails that have lost their distinct line-shaped appearance. Ground-based cameras can help fill this gap by capturing high-resolution images of contrails during their early stages and ensuring more detail observation of when they are formed, providing essential verification data for more accurate climate predictions.

Read the full research paper

“We’re excited to provide ADS-B based flight data to help build benchmark datasets, supporting the research community in enhancing current contrail and weather models. This collaboration showcases the incredible work by the Imperial College research team and reinforces our commitment to fostering industry-research partnerships for innovative aviation sustainability solutions.”

Johan Alex Varghese
Head of Commercial & Partnerships, Spire Aviation

Share
Share on LinkedIn

Ground-based cameras and ADS-B data: A step forward in validation contrail simulations

In a recent study, scientists at Imperial College London leveraged automatic dependent surveillance-broadcast (ADS-B) data from Spire Aviation, historical weather information, and contrail models to compare simulated contrails with those captured by ground-based cameras. The data was collected in Central London using a Raspberry Pi camera between October 2021 and April 2022. After excluding footage with low-level clouds or poor visibility, they analyzed 14 hours of video from five different days. ADS-B data and simulated contrail dimensions were overlaid on the video (see video below) to compare the formation, lifetime, and width of observed contrails.


Example of the ADS-B telemetries and simulated contrail segments that are superimposed to the video footage.

Comparing simulated and observed contrails: Insights from ADS-B data and weather models

A total of 1,582 waypoints from 281 flights were identified in the footage. The simulation accurately predicted contrail formation and their absence for about 75% of the waypoints, with errors occurring more often in warmer temperatures. Among the contrails observed, 73% had short lifetimes of less than two minutes, while 14% persisted for over 10 minutes.

Weather data from numerical prediction models generally indicated that short-lived contrails formed when the air was too dry to maintain ice (relative humidity with respect to ice, RHi < 100%), whereas long-lasting contrails were more likely when the air was ice-supersaturated (RHi > 100%).

“Although we only looked at a small number of contrails formed under clear sky conditions, our results show that current simulation tools can accurately predict contrail formation around 75% of the time. Additionally, existing weather models are generally capable of explaining why some contrails persist longer than others.”

Dr. Roger Teoh
Honorary Research Fellow – Imperial College London

Share
Share on LinkedIn

On average, simulated contrail widths were approximately 100 meters narrower than those observed, with the largest discrepancies occurring in the first five minutes after formation. The differences between observed and simulated contrail properties may be due to uncertainties in weather data, aircraft performance estimates, simplifications in the contrail models, and challenges in detecting faint or overlapping contrails with natural clouds, among other factors.

Ground-based observations and ADS-B data: Key to advancing contrail and weather models

This study highlights how ground-based observations can be effectively used to evaluate the accuracy of weather and contrail models. Despite the small sample size and the focus on contrails formed in clear sky conditions, the results indicate that current contrail simulation tools accurately predict contrail formation around 75% of the time. Furthermore, existing weather models provide valuable insights into why some contrails last longer than others. The ADS-B telemetry data and contrail observations from this study offer an important benchmark data set that will help researchers refine, validate, and compare different weather and contrail models moving forward.

Unlock the power of space data to solve sustainability challenges

More Blog Posts