SatCAT: Using satellite data to improve safety and routing efficiency in the aviation industry.

Lead Research Organisation: University of Oxford
Department Name: Oxford Physics

Abstract

Turbulence is an important safety consideration for commercial air transport: Crashes caused by turbulence are exceedingly rare, but injuries to passengers and crew are a weekly occurrence. It can often strike without warning, particularly "clear air turbulence" that is very difficult to detect, meaning that the seat belt sign may be switched off and people will be moving about the cabin. In this situation, people can be thrown off their feet, injuring themselves in the process, and any loose items (such as laptops) can become hazardous projectiles.

Airlines and pilots therefore plan routes that avoid potential areas of turbulence and - once in the air - will alter course whenever possible if they suspect severe turbulence is ahead. This, however, relies on adequate warning being available: It is impossible to avoid turbulence if you do not know it is there. Weather forecasts provide turbulence warnings, but these are often inaccurate - showing turbulence where none exists or failing to forecast turbulence when it does exist. Similarly, it is very hard to detect turbulence from the cockpit. Clear air turbulence is invisible and storm-associated turbulence is hard to decipher. A particular challenge is to work out by how far an aircraft needs to avoid storm clouds in order to remain clear of turbulence.

In this fellowship, I aim to increase our knowledge of turbulence in order to help to provide new guidance to pilots on how to avoid storm turbulence. I will produce a new method that will detect turbulence from space using the latest weather satellite technology, enabling better flight planning and helping airlines to avoid turbulence: Keeping their passengers safe and saving money by reducing the need for unnecessary flight plan changes around non-existent turbulence. By working with weather forecasting experts at the Met Office and University of Reading I will seek to improve our ability to forecast turbulence, allowing airlines not only to avoid turbulence in-flight but also to allow better advance planning of turbulence-free routes before take-off.
 
Description Discovered extremely cold thunderstorm in Tropical Pacific, which tested capabilities of satellite sensor technology. My work examined how we can measure such storms accurately and also looked at whether these 'extreme' storms are becoming more frequent.

Used techniques developed during this fellowship to study the Hunga-Tonga volcanic eruption. We found that the debris from this eruption was lofted into the mesosphere, the first recorded instance of this happening and likely the first time it has happened since Krakatoa.
Exploitation Route Important for climate modelling, which underestimates frequency of such storms.

New information about volcanic eruptions, which can be used to examine climatic influence of both previous and future volcanic events.
Sectors Aerospace, Defence and Marine,Agriculture, Food and Drink,Energy,Environment

 
Description Some my results are being used by EUROCONTROL, the EU-wide coordinator of air traffic control services, to map when and where severe thunderstorms are most likely to occur across their control regions. This work enables them to engage in strategic planning to counteract the threat posed by weather to aircraft by, for example, providing additional air traffic capacity in regions likely to be affected by thunderstorms (thus allowing weather avoidance trajectories to be conducted). This has the primary effect of reducing delays caused by regional congestion (a significant problem over, for instance, the Alps), allowing economic benefit in terms of reduced fuel burn, fewer delays and more predictable scheduling. I have also been working with various airlines (who, unfortunately, prefer not to be named) in analysing serious weather-related incidents that have occurred on their aircraft. This work helps understand what happened and, when necessary, adjust training and procedures to prevent such incidents happening again, an important societal impact in terms of safety. In early 2020 I began a collaboration with TUI, the world's largest charter airline, to jointly research how climate change will affect their operations and to devise mitigation strategies. COVID has now put much of the industrial work on standby, as most of my airline colleagues are furloughed. However, new links have been forged with the Omani Directorate General of Meteorology to explore improved forecasting of thunderstorms, which is showing positive initial results.
First Year Of Impact 2019
Sector Aerospace, Defence and Marine,Transport
Impact Types Societal,Economic

 
Description N/A
Geographic Reach Europe 
Policy Influence Type Implementation circular/rapid advice/letter to e.g. Ministry of Health
 
Description ASSOCIATE SCIENTIST MISSION FOR THE EUMETSAT SATELLITE APPLICATION FACILITY ON NUMERICAL WEATHER PREDICTION (NWP SAF)
Amount € 14,732 (EUR)
Organisation Meteorological Office UK 
Sector Academic/University
Country United Kingdom
Start 04/2020 
End 06/2020
 
Description SatFlood: Providing early-warning of flash flood events in Kenya
Amount £13,591 (GBP)
Organisation University of Oxford 
Sector Academic/University
Country United Kingdom
Start 02/2019 
End 07/2019
 
Title Go-around detector 
Description A tool to detect aircraft go-arounds in any form of aircraft position data, tested with OpenSky data at VABB airport. 
Type Of Material Improvements to research infrastructure 
Year Produced 2020 
Provided To Others? Yes  
Impact None as yet 
URL https://github.com/simonrp84/Go_Around_Detector
 
Description Air traffic control in poor weather / turbulence 
Organisation National Office for Aerospace Studies and Research
Country France 
Sector Public 
PI Contribution Provision of predicted clear-air turbulence location data for use in a study of European air transport management.
Collaborator Contribution Provision of aircraft and air traffic control data to the study.
Impact Too early for outcomes, project still active.
Start Year 2019
 
Description Airline study of thunderstorm activity 
Organisation EasyJet
Country United Kingdom 
Sector Private 
PI Contribution Creation of a large database of thunderstorm events for use by the airline in assessing the historical risk posed to their aircraft. Analysis of specific case studies in which a thunderstorm has generated a risk event for a particular flight.
Collaborator Contribution Supply of aircraft and route data, supply of flight data recorder information. Guidance and support.
Impact Internal publications on thunderstorm risk,.
Start Year 2018
 
Description Preparations for upcoming new European satellite missions 
Organisation European Organisation for the Exploitation of Meteorological Satellites
Country Germany 
Sector Public 
PI Contribution EUMETSAT requested external experience for testing the planned data to be sent back by upcoming satellite missions (due to launch 2022). I used example data within my research, exploring how it can be used for storm detection, ash cloud detection and land surface analysis. This was fed back to EUMETSAT to enable them to tune their data portfolio and algorithms.
Collaborator Contribution EUMETSAT provided confidential sample data for the upcoming missions.
Impact Adjustments to products developed by EUMETSAT, which will have a direct impact on weather and climate monitoring over Europe and Africa once the new satellites are launched.
Start Year 2020
 
Description RAL Space / STFC collaboration for ORAC 
Organisation Rutherford Appleton Laboratory
Department RAL Space
Country United Kingdom 
Sector Academic/University 
PI Contribution Adding and updating computer code that forms the ORAC cloud and aerosol retrieval algorithm.
Collaborator Contribution Additional ORAC code, plus accuracy validation of the output data to ensure it is suitable for use. Presentation of ORAC at conferences on my behalf.
Impact None as yet, ongoing.
Start Year 2015
 
Description SatPy: Satellite data processing toolkit 
Organisation European Organisation for the Exploitation of Meteorological Satellites
Country Germany 
Sector Public 
PI Contribution Additions to the 'Satpy' meteorological data processing library (a interface for python scripts) to allow reading of data produced by new satellites (such as Fengyun-4) and new data types (such as cloud properties from EUMETSAT's NWC-SAF).
Collaborator Contribution Substantial technical support to enable me + my colleagues at Uni. Oxford to work with the data we need in Satpy. Substantial programming time spend adding new methods, tools and interfaces to satpy to enable it to be better suited to research use.
Impact None yet
Start Year 2018
 
Description SatPy: Satellite data processing toolkit 
Organisation Norwegian Meteorological Institute
Country Norway 
Sector Public 
PI Contribution Additions to the 'Satpy' meteorological data processing library (a interface for python scripts) to allow reading of data produced by new satellites (such as Fengyun-4) and new data types (such as cloud properties from EUMETSAT's NWC-SAF).
Collaborator Contribution Substantial technical support to enable me + my colleagues at Uni. Oxford to work with the data we need in Satpy. Substantial programming time spend adding new methods, tools and interfaces to satpy to enable it to be better suited to research use.
Impact None yet
Start Year 2018
 
Description SatPy: Satellite data processing toolkit 
Organisation Swedish Meteorological and Hydrological Institute
Country Sweden 
Sector Academic/University 
PI Contribution Additions to the 'Satpy' meteorological data processing library (a interface for python scripts) to allow reading of data produced by new satellites (such as Fengyun-4) and new data types (such as cloud properties from EUMETSAT's NWC-SAF).
Collaborator Contribution Substantial technical support to enable me + my colleagues at Uni. Oxford to work with the data we need in Satpy. Substantial programming time spend adding new methods, tools and interfaces to satpy to enable it to be better suited to research use.
Impact None yet
Start Year 2018
 
Description SatPy: Satellite data processing toolkit 
Organisation University of Wisconsin-Madison
Country United States 
Sector Academic/University 
PI Contribution Additions to the 'Satpy' meteorological data processing library (a interface for python scripts) to allow reading of data produced by new satellites (such as Fengyun-4) and new data types (such as cloud properties from EUMETSAT's NWC-SAF).
Collaborator Contribution Substantial technical support to enable me + my colleagues at Uni. Oxford to work with the data we need in Satpy. Substantial programming time spend adding new methods, tools and interfaces to satpy to enable it to be better suited to research use.
Impact None yet
Start Year 2018
 
Description Storm cloud properties analysis 
Organisation Monash University
Country Australia 
Sector Academic/University 
PI Contribution Improvements made to the ORAC cloud properties retrieval algorithm to enable optimal use with geostationary satellites covering Australia
Collaborator Contribution Research into the properties of storms using tools developed as part of my fellowship, joint analysis of the results. Essentially, Monash provided person-time to help analyse data.
Impact Not yet relevant
Start Year 2019
 
Title Automated Go-Around detector 
Description A tool to detect aircraft go-arounds in any form of aircraft position data, tested with OpenSky data at VABB airport. 
Type Of Technology Software 
Year Produced 2019 
Open Source License? Yes  
Impact None as yet 
URL https://github.com/simonrp84/Go_Around_Detector
 
Title ORAC: Optimal Retrieval of Aerosol and Cloud 
Description ORAC-CC4CL (the Optimal Retrieval of Aerosol and Cloud - Community Code for CLimate) is an optimal estimation retrieval scheme for the estimation of aerosol and cloud properties from visible-infrared imaging satellites such as MODIS, AATSR, AVHRR and SEVIRI. 
Type Of Technology Software 
Open Source License? Yes  
Impact From me: Application of this software within the aviation industry, providing a database of historical cloud information for use by airline climate resilience teams. 
URL https://github.com/ORAC-CC/orac/wiki
 
Title pytroll/satpy: Version 0.34.0 (2022/02/18) 
Description Issues Closed Issue 2026 - Missing units in avhrr_l1b_eps reader (PR 2027 by @gerritholl) Issue 2024 - Allow to skip unit conversion in ninjotiff writer (PR 2025 by @gerritholl) Issue 2023 - Allow to keep units in composite Issue 2022 - save_dataset changes dataset in-place Issue 2018 - Wrong AxisIntercept (add_offset) when writing °C temperature units with ninjogeotiff writer Issue 2014 - Problem in converting VIIRS hdf to geotif Issue 2010 - AHI HSD true_color incorrect with cache_sensor_angles (PR 2013 by @djhoese) Issue 2008 - abi_l1b reader leaks memory in Python-3.7 (PR 2011 by @sfinkens) Issue 2004 - Configure image type returned by MaskingCompositor (PR 2005 by @gerritholl) Issue 2001 - Failed to load AVHRR LAC data Issue 1999 - Reader for ???????-? (Arktika-M) ???-?? (MSU-GS) data (PR 2000 by @simonrp84) Issue 1998 - Add reader for Arctica M N-1 hdf5 data Issue 1995 - AttributeError when cropping data for VIIRS Issue 1959 - Unittest failure in test_modifiers.py Issue 1948 - Contribute to Satpy Issue 1945 - Wrong dtype of uint32 array saved by the cf_writer Issue 1943 - sza_check from trollflow2 fails with KeyError: 'start_time' Issue 1883 - Test failure on i386 and armhf (PR 1966 by @djhoese) Issue 1384 - AHI HRIT reader has gotten slower (PR 1986 by @pnuu) Issue 1099 - find_files_and_readers read unneeded files In this release 20 issues were closed. Pull Requests Merged Bugs fixed PR 2027 - Include units with AVHRR EPS metadata (2026) PR 2017 - Fix ABI rayleigh_corrected_crefl modifier using deprecated DEM specifier PR 2015 - Fix various dask array bugs in CREFL modifier PR 2013 - Fix angle generation caching occassionally swapping results (2010) PR 2011 - Fix memory leak in cached_property backport (2008, 2008) PR 2006 - Fix Scene not being serializable PR 2002 - Update tests to be more flexible to CRS and enhancement changes PR 1991 - Update reference to dask distributed setup page PR 1988 - Update geometry.py docstring from compositor to modifier PR 1987 - Check that time is not already a coordinate in CF writer PR 1983 - More general filename filter for ascat soil moisture, allowing for Metop-B and Metop-C PR 1982 - Fix ninjotiff writer from erraneous K to C conversion Features added PR 2025 - Allow skipping unit conversion in NinJoTIFF (2024) PR 2007 - Update abi_l2_nc to include filename metadata similar to abi_l1b PR 2005 - Add flag to MaskingCompositor to return RGBA for single-band input (2004) PR 2000 - Add a reader for the MSU-GS/A + Arctica-M1 data (1999) PR 1992 - Add support for CMIC product from PPSv2021 PR 1989 - read the "elevation" variable in slstr_l1b PR 1986 - Add reader kwarg to 'ahi_hrit' to disable exact start_time (1384) PR 1967 - Add ability to read comma-separated colormaps during enhancement PR 1966 - Reduce MODIS L1b/L2 test case size for better test performance (1883) PR 1962 - Use a dependency matrix for benchmarking Documentation changes PR 2020 - Clarify documentation regarding attributes used in get_angles PR 1991 - Update reference to dask distributed setup page PR 1988 - Update geometry.py docstring from compositor to modifier PR 1969 - Improve modifier documentation PR 1968 - Improve API documentation in CompositeBase PR 1961 - Update documentation to refer to all EO satellite data PR 1960 - Add release notes and security policy to documentation PR 1950 - Fix formatting in configuration documentation In this release 30 pull requests were closed. 
Type Of Technology Software 
Year Produced 2022 
Open Source License? Yes  
Impact Used at meteorological agencies around the world, primary method for working with satellite data at the Swedish, German, Danish, Finnish, Slovenian, Australian and Singapore met offices. Used operationally by companies, military and other government entities. 
URL https://zenodo.org/record/6149097
 
Description Media engagement regarding Tonga volcano eruption 
Form Of Engagement Activity A press release, press conference or response to a media enquiry/interview
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Media (as a channel to the public)
Results and Impact Received multiple requests for interview as a result of Tonga volcano eruption on 15th Jan 2022. Mainly print / online media but also radio + TV to a lesser extent.
Year(s) Of Engagement Activity 2022
 
Description Multiple interviews and media discussions regarding meteor impact 
Form Of Engagement Activity A press release, press conference or response to a media enquiry/interview
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Media (as a channel to the public)
Results and Impact I published on a social media feed a short video using satellite data that showed the impact of a large meteor off the coast of Alaska. This was picked up by journalists, initially the BBC, which led to many media enquiries from both UK and foreign outlets and included requests for quotes, background interviews and use of the original video + images.

My contact at the BBC reported that their page relating to this event received over 7 million unique page views, one of the most read news articles of the month, and analysis by my employer suggests at least 30 million unique pageviews across all media outlets.
Year(s) Of Engagement Activity 2019
URL https://www.bbc.com/news/science-environment-47607696
 
Description Press release and media engagement for new paper 
Form Of Engagement Activity A press release, press conference or response to a media enquiry/interview
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Media (as a channel to the public)
Results and Impact Press release for a new paper that sparked considerable media interest, including interviews for various radio and TV broadcasts as well as print/online news articles.
Year(s) Of Engagement Activity 2021
URL https://www.bbc.co.uk/news/science-environment-56542408
 
Description Social media channel 
Form Of Engagement Activity Engagement focused website, blog or social media channel
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Public/other audiences
Results and Impact I run a personal twitter channel that focuses on my research, mainly tweeting examples of satellite data or discussing their relevence to weather forecasting or the aviation industry. Median monthly 'reach' is ~50,000 unique engagements. Maximum monthly reach over 2019 was 2.3 million unique engagements.
Year(s) Of Engagement Activity 2016,2017,2018,2019,2020,2021,2022