ABSTRACT A simple self-contained theory is proposed for describing life cycles of convective syst... more ABSTRACT A simple self-contained theory is proposed for describing life cycles of convective systems as a discharge-recharge process. A closed description is derived for the dynamics of an ensemble of convective plumes based on an energy cycle. The system consists of prognostic equations for the cloud work function and the convective kinetic energy. The system can be closed by introducing a functional relationship between the convective kinetic energy and the cloud-base mass flux. The behaviour of this system is considered under a bulk simplification. Previous cloud-resolving modelling as well as bulk statistical theories for ensemble convective systems suggest that a plausible relationship would be to assume that the convective kinetic energy is linearly proportional to the cloud-base mass flux. As a result, the system reduces to a nonlinear dynamical system with two dependent variables, the cloud-base mass flux and the cloud work function. The fully nonlinear solution of this system always represents a periodic cycle regardless of the initial condition under constant large-scale forcing. Importantly, the inclusion of energy dissipation in this model does not in itself lead the system to an equilibrium.
ABSTRACT Forecasting centres routinely run simulations at convection-permitting resolutions, but ... more ABSTRACT Forecasting centres routinely run simulations at convection-permitting resolutions, but there is an urgent need for novel radar-observation techniques to evaluate the storm structures produced by these models. A data set of high-resolution radar observations for forty days with convective storms is used to evaluate such storms in the UK Met Office forecast model for the DYMECS project (Dynamical and Microphysical Evolution of Convective Storms). The 3 GHz Chilbolton radar was set up to automatically track convective storms in real-time through a scan-scheduling algorithm linked to a database of storms identified in the Met Office rainfall radar network. Many configurations of the Met Office model have been tested against the Chilbolton observations for their representation of convective storms. In terms of the detailed three-dimensional microphysical structure, modelled storms are shown to generally have wider horizontal structures for different reflectivity thresholds compared to radar observations, whilst the storm cores are not as deep as observed. For instance, the model does not produce reflectivities above 40 dBZ above the melting layer, which are frequently observed in intense convective storms, but this is improved by the inclusion of prognostic graupel in the microphysics scheme. The dynamical storm structures are analysed in terms of vertical velocity and the size of convective cores, derived from vertical profiling radar scans. Two existing retrieval methods for vertical velocities are presented and compared statistically. Both methods are then combined with radar reflectivity observations to define convective cores and to evaluate the size and intensity of such cores in the model. The results presented here will help improve the microphysics and sub-grid mixing schemes, as well as determine whether even higher resolution (down to 100m) models provide a notably better representation of the three-dimensional evolution of convective storms.
The theoretical arguments and practical justifications are becoming well established for the intr... more The theoretical arguments and practical justifications are becoming well established for the introduction of some stochastic component(s) to the physical parameterisations used by weather forecast and climate models. For example, many general-circulation models are known to have have insufficient high-frequency, small-scale variability of convective heating rates and precipitation in the tropics, which may damage their ability to represent low-frequency, large-scale climate variability. A wide variety of stochastic methods have been investigated, ranging from the simple (e.g. varying some uncertain model parameter) through to the more complicated, which may attempt to deal directly with the physical mechanisms that lead to variability near to the model gridscale. I will outline one of the more complex methods (which recognizes that the number of convective elements in a grid box need not be large) and discuss results from this and some simpler methods in order to address the question posed in the title. Or, in other words, might the results obtained from more complicated methods ultimately provide a justification for the use of simpler approaches in practice? Alternatively, is a generic inclusion of noise sufficient or does the physical origin of the small-scale variability impose any necessary or desirable constraints on the character of the noise?
ABSTRACT The leading cause of flash floods in the United Kingdom is severe convection. The predic... more ABSTRACT The leading cause of flash floods in the United Kingdom is severe convection. The prediction of such convection poses a challenge to the state of the art numerical weather prediction because error doubling times are short (hours) compared to the lead times necessary for useful warnings. The predictability of convection also changes from case to case. Here we examine the predictability of two different convective events over the UK. The first is a case of widespread convection largely forced by the large scale, and the second is a series of relatively small convective cells which moved over the same small river catchment and caused a flash flood. The Unified Model of the Met Office is run with 1 km grid spacing and ensemble members are generated perturbing the potential temperature every 30 minutes as the simulation progresses. The predictability is addressed by comparison of the accumulations over the duration of the events as simulated by the different ensemble members and the different characteristics of the effects of the perturbations are determined. Preliminary results show that the intensity of the perturbation is the main factor in determining the error growth rate. Furthermore, for small perturbations, there is indication of a sensitivity of the growth rate to the time of day.
ABSTRACT A simple self-contained theory is proposed for describing life cycles of convective syst... more ABSTRACT A simple self-contained theory is proposed for describing life cycles of convective systems as a discharge-recharge process. A closed description is derived for the dynamics of an ensemble of convective plumes based on an energy cycle. The system consists of prognostic equations for the cloud work function and the convective kinetic energy. The system can be closed by introducing a functional relationship between the convective kinetic energy and the cloud-base mass flux. The behaviour of this system is considered under a bulk simplification. Previous cloud-resolving modelling as well as bulk statistical theories for ensemble convective systems suggest that a plausible relationship would be to assume that the convective kinetic energy is linearly proportional to the cloud-base mass flux. As a result, the system reduces to a nonlinear dynamical system with two dependent variables, the cloud-base mass flux and the cloud work function. The fully nonlinear solution of this system always represents a periodic cycle regardless of the initial condition under constant large-scale forcing. Importantly, the inclusion of energy dissipation in this model does not in itself lead the system to an equilibrium.
ABSTRACT Forecasting centres routinely run simulations at convection-permitting resolutions, but ... more ABSTRACT Forecasting centres routinely run simulations at convection-permitting resolutions, but there is an urgent need for novel radar-observation techniques to evaluate the storm structures produced by these models. A data set of high-resolution radar observations for forty days with convective storms is used to evaluate such storms in the UK Met Office forecast model for the DYMECS project (Dynamical and Microphysical Evolution of Convective Storms). The 3 GHz Chilbolton radar was set up to automatically track convective storms in real-time through a scan-scheduling algorithm linked to a database of storms identified in the Met Office rainfall radar network. Many configurations of the Met Office model have been tested against the Chilbolton observations for their representation of convective storms. In terms of the detailed three-dimensional microphysical structure, modelled storms are shown to generally have wider horizontal structures for different reflectivity thresholds compared to radar observations, whilst the storm cores are not as deep as observed. For instance, the model does not produce reflectivities above 40 dBZ above the melting layer, which are frequently observed in intense convective storms, but this is improved by the inclusion of prognostic graupel in the microphysics scheme. The dynamical storm structures are analysed in terms of vertical velocity and the size of convective cores, derived from vertical profiling radar scans. Two existing retrieval methods for vertical velocities are presented and compared statistically. Both methods are then combined with radar reflectivity observations to define convective cores and to evaluate the size and intensity of such cores in the model. The results presented here will help improve the microphysics and sub-grid mixing schemes, as well as determine whether even higher resolution (down to 100m) models provide a notably better representation of the three-dimensional evolution of convective storms.
The theoretical arguments and practical justifications are becoming well established for the intr... more The theoretical arguments and practical justifications are becoming well established for the introduction of some stochastic component(s) to the physical parameterisations used by weather forecast and climate models. For example, many general-circulation models are known to have have insufficient high-frequency, small-scale variability of convective heating rates and precipitation in the tropics, which may damage their ability to represent low-frequency, large-scale climate variability. A wide variety of stochastic methods have been investigated, ranging from the simple (e.g. varying some uncertain model parameter) through to the more complicated, which may attempt to deal directly with the physical mechanisms that lead to variability near to the model gridscale. I will outline one of the more complex methods (which recognizes that the number of convective elements in a grid box need not be large) and discuss results from this and some simpler methods in order to address the question posed in the title. Or, in other words, might the results obtained from more complicated methods ultimately provide a justification for the use of simpler approaches in practice? Alternatively, is a generic inclusion of noise sufficient or does the physical origin of the small-scale variability impose any necessary or desirable constraints on the character of the noise?
ABSTRACT The leading cause of flash floods in the United Kingdom is severe convection. The predic... more ABSTRACT The leading cause of flash floods in the United Kingdom is severe convection. The prediction of such convection poses a challenge to the state of the art numerical weather prediction because error doubling times are short (hours) compared to the lead times necessary for useful warnings. The predictability of convection also changes from case to case. Here we examine the predictability of two different convective events over the UK. The first is a case of widespread convection largely forced by the large scale, and the second is a series of relatively small convective cells which moved over the same small river catchment and caused a flash flood. The Unified Model of the Met Office is run with 1 km grid spacing and ensemble members are generated perturbing the potential temperature every 30 minutes as the simulation progresses. The predictability is addressed by comparison of the accumulations over the duration of the events as simulated by the different ensemble members and the different characteristics of the effects of the perturbations are determined. Preliminary results show that the intensity of the perturbation is the main factor in determining the error growth rate. Furthermore, for small perturbations, there is indication of a sensitivity of the growth rate to the time of day.
Uploads
Papers by Robert Plant