Dr Neill Bowler
Neill is a scientist working on the use of GNSS radio-occultation observations in weather forecasting.
Areas of expertise
- Use and assimilation of GNSS radio occultation data
- Ensemble prediction (particularly their initial conditions)
- Data assimilation (particularly the use of ensemble covariances)
- Verification of ensemble forecasts
- Diagnosing model errors
Current activities
Neill’s current work focusses on GNSS radio occultation (GNSS-RO). GNSS (Global Navigation Satellite Systems, such as GPS) can be used to provide observations of the earth's atmosphere. If a second satellite receives a signal from the GNSS network, sometimes that signal will have skimmed the earth's atmosphere. The amount of refraction of the signal depends on the density and amount of water vapour in the atmosphere. So, by measuring the total refraction it is possible to estimate some atmospheric properties. Neill's work focuses on the use of these observations to improve the weather forecasting system.
Neill is currently working on the monitoring of new data, and its use in the data assimilation system. A particular interest is in the impact of adding new observations to the numerical weather prediction system. He is also looking at the estimates of errors used in these observations.
Career background
Neill started work for the Met Office in 2001, looking at the precipitation nowcasting system. He was heavily involved in developing the ensemble prediction system for precipitation nowcasts. In 2003, Neill moved to join ensemble forecasting research, followed by the ensemble data assimilation group. Since 2017 Neill has started work on the use of GNSS radio occultation observations.
Previous work has focused on developing an ensemble of data assimilations. This system is based on the four-dimensional variational data assimilation method used operationally for assimilating observations. However, it does not require the expensive tangent-linear model which makes it very useful for ensemble generation. Being close in method to the data assimilation system should mean it is effective at simulating the uncertainties in the data assimilation, which will improve the weather forecast.
Ensemble forecasts are produced by running the forecast model a number of times, with slightly different starting conditions and small perturbations to the forecast model. Neill's main activity in this area is to develop the capability of the new system to generate initial conditions for ensemble forecasts. In particular, he looked at how to inflate the ensemble perturbations so that they fully cover the range of possible forecast outcomes.
Neill has also worked on the verification of forecasts, which involves comparing past forecasts with observations to assess how skilful the forecasts were. Neill's work covered a number of novel verification methods, particularly for probabilistic forecasts. Neill provided support for the ensemble prediction system used by other national services through the Unified Model collaboration.
Prior to joining the Met Office, Neill completed a PhD in Physics at the University of Warwick. This work was on the role of noise in optimisation problems and in diffusion-limited aggregation.