This School aims to educate and motivate young talented PhD students and early career scientists through three specialist lecture courses (supplemented by tutorials) and guest lectures that will provide a panorama of some of the most promising research areas within this interdisciplinary field. The school is organised in partnership with the London Mathematical Society and with the support of the EPSRC-funded CDT Mathematics of Planet Earth (Imperial/Reading).

Speakers (See abstracts at the bottom of this page):
– Course 1: Hakima Bessaih (University of Wyoming): Data-assimilation of PDEs through nudging and Random vortex methods.
– Course 2: Alberto Carrassi (NERSC, Norway): Mathematical Theory of Data Assimilation with Applications
– Course 3: Darryl Holm (Imperial College, London): Variational principles for Stochastic Fluid Dynamics

Guest Lectures:
– Beatrice Pelloni (Heriot-Watt): Mathematical aspects of the semigeostrophic system
– Peter Ashwin (Exeter): Response of the Pleistocene ice-ages to astronomical forcing

Tutors:
Course 1: Giulia Carigi (Reading); Course 2: Colin Grudzien (Assistant Professor of Statistics, University of Nevada, Reno); Course 3: Erwin Luesink (Imperial)

Guest Speaker at visit to MERL:
– Ros Cornforth (Walker Institute, Reading): Climate Information for Decision-Making: Responding to Climate Change in Developing Countries.

Speaker Abstracts:

Hakima Bessaih (University of Wyoming): Data-assimilation of PDEs through nudging and Random vortex methods.

An introduction to some data assimilation methods for PDEs are introduced. They are mainly based on nudging techniques that are motivated by feedback controllers. The observational measurements are given by finite volumes or nodal points and some random errors in the observations are taken into account. We analyze the performance of such algorithms. Specific bounds are given in terms of the resolution density of the observations and the parameters of the systems studied. Results on the average time errors in mean are stated.  The last lecture will be devoted to the introduction of some random vortex methods for 3d fluids. Specifically, some dynamics of vortex filaments will be stated along with a mean field and propagation of chaos results. Their connections to 3d vector valued PDEs and in particular 3d Euler equation will be analyzed.

Alberto Carrassi (NERSC, Norway): Mathematical Theory of Data Assimilation with Applications

State estimation theory in geosciences is commonly referred to as data assimilation. This term encompasses the entire sequence of operations that, starting from the observations of a system, and from additional statistical and/or dynamical information (such as an evolution model), provides an estimate of its state. Data assimilation is common practice in numerical weather prediction but its application is becoming widespread in many other areas of climate, atmosphere, ocean and environment modelling. The course will provide first the formulation of the problem from a Bayesian perspective and will then present the two popular families of Gaussian based approaches, the Kalman-filter/-smoother and the variational methods. Ensemble based methods will then be considered, starting from the well-known Ensemble Kalman filter, in its stochastic or deterministic formulation, and then the state-of-the-art ensemble-variational methods, as well as particle filters. The course will focus on the specific challenges that data assimilation has encountered to deal with high-dimensional chaotic systems, such as the atmosphere and ocean, and the countermeasures that have been taken and which have driven the recent dramatic development of the field.

Darryl Holm (Imperial College, London): Variational principles for Stochastic Fluid Dynamics

The lectures will be based on Hamilton’s principle for ideal fluids and its infinite dimensional symmetries for temporally stochastic evolution of fluids under spatially smooth invertible maps. This means that the Lagrangian particle trajectories exist as stochastic paths generated by Eulerian stochastic vector fields. Thus, Hamilton’s principle retains its original gauge symmetry under relabelling of Lagrangian particles. Because this gauge symmetry persists, Noether’s theorem delivers Kelvin’s theorem for circulation dynamics of the stochastic flow in the same form as for the deterministic case, but now Kelvin’s circulation loop moves with the fluid along a stochastic path. The corresponding nonlinear stochastic PDEs for the fluid motion provide a probabilistic estimation of model error, based on the observed spatial correlations of uncertainties in the fluid transport. Thus, this variational approach yields a data-driven model of stochastic flow for probabilistic estimates of Variability in Geophysical Fluid Dynamics (GFD). Details and examples for GFD may be found in: D.D. Holm, Proc Roy Soc A, 471: 20140963 (2015)

Arrival & Registration

Sunday dinner and Monday Breakfast: Dinner is served between 18.00 – 19.45, Breakfast between 07.30 – 09.00 at Eat at the Square (no vouchers necessary as your name will be on a list at the tills).

Monday morning: Registration in Lyle G74 from 08.30 – 09.00 – you will receive your meal & drinks vouchers and badges at this time.

Timetable:

Lecture Rooms:

The room Lyle G74 is located in the Phillip Lyle (building 32 on the map). It is in the building on the ground floor. You go up the stairs (ramp) enter the building and turn left. The lecture hall GU01 is located in the meteorology building (building 58 on the map), on the ground floor.

Organisers:
Valerio Lucarini,Tobias Kuna, Jennifer Scott, Dan Crisian
Email the organizers at t.kuna-AT-reading.ac.uk.

Funding:
The school is in partnership with the London Mathematical Society (LMS). and supported by the EPSRC Centre of Doctoral Training Mathematics of Planet Earth. The LMS is the UK’s learned society for mathematics. Registered charity no. 252660 (www.lms.ac.uk)

Travelling to Reading Find out how to reach us from the airport, by train, and from Reading town centre here: http://www.reading.ac.uk/about/visit-us.aspx

Arriving by Air

• From Heathrow: A direct bus service runs between Heathrow airport and Reading station. Buses run every 30 minutes, during the day, and take about 40 minutes. For more information, please visit https://www.firstgroup.com/berkshire-thames-valley/rail-air/about You can get a ticket either from the driver or in advance online https://www.firstgroup.com/berkshire-thames-valley/rail-air/buy-tickets
  • Arrivals in Terminals 2 & 3: follow the signs for bus and coach, this takes you underground, then follow the signs Heathrow Central Bus Station, a 4 – 7 minute walk.
  • Arrivals in Terminal 5: follow signs for Local Buses and go to Stand 9.
  • Arrivals in Terminal 4: use the FREE Heathrow Express rail shuttle service to the Heathrow Central Bus Station at Terminals 2 & 3.
• From Gatwick: there are direct trains from Gatwick Airport (GTW) to Reading Station (RDG) approximately once an hour during the day and takes about 75 minutes. You can buy tickets at the station or online https://www.nationalrail.co.uk/ If you wish to book a taxi from either Heathrow or Gatwick to Whiteknights Campus contact one of the University approved taxi companies for a quote:
  • YellCars: http://yellcars.com/ or email manager@yellcars.com
  • Loddon Cars: https://www.loddoncars.com/ or email bookings@loddoncars.com

  A taxi from Heathrow will cost from approximately £50 depending on the number of passengers and items of luggage. Parking and waiting time will be charged extra. A taxi from Gatwick will cost from approximately £90 depending on the number of passengers and items of luggage. Parking and waiting time will be charged extra.

Local Transport 

Once you arrive at the train station you can take a taxi or bus, the bus stops and taxi rank can be found outside the station.

• A taxi will cost you approximately £8 and a single bus journey will cost you £2.00 (a day ticket is £4.00. You purchase your ticket on the bus and will need either the exact change or a contactless bank card).
• Buses: https://www.reading-buses.co.uk/
• A comprehensive guide to finding the University: http://www.reading.ac.uk/about/visit-us.aspx
• Reading Station information: https://www.networkrail.co.uk/stations/reading/
• The station map might help you orient yourself: https://cdn.networkrail.co.uk/wp-content/uploads/2019/04/Reading-Station-Map.pdf

Maps

• A map of the town
• A map of the campus with the main venues of the school
• Google map location of the Mathematics and Statistics Department
• Google map location of the Meteorology department were the lectures take place.

Housing 

All housing are on the campus.

• Students will have accommodations in students housing: Padworth HouseYou will receive relevant details by email. See also the map
• Lecturers will have accommodations in: The Cedars hotel. See also the map

Arrival instructions: you will receive details from Venue Reading nearer the date, but Check-in for accommodation is usually from 2pm, and check out at 10am on the day of departure. Late arrivals can use the call point on the wall outside reception which will contact security who will issue the room key.

Social Activity MERL:

We have a social activity on Wednesday 10th July. We will have barbecue at MERL (Museum for English Rural Life. The museum is in walking distance from the campus: 6 Redlands Rd, Reading RG1 5EX (google maps link) see also the following map and on the map of the town

Catering

Breakfast and Dinner are self-service Conference meals, delegates names are checked off a list.

Evening meals: self-service hot conference dinner at Eat at the Square from Sunday evening through Thursday evening, except for the BBQ on Wednesday, voucher for drinks which can be used at Park House – a bottled beer/a glass of wine/a soft drink.

Dinner choice comprises:

• 2 meat hot dishes
• 1 fish hot dish
• 1 vegetarian dish
• Rice/potatoes/pasta as appropriate
• Seasonal vegetables
• Salad bar
• Bread rolls with butter
• Fresh fruit salad
• Dessert
• Tea & Coffee

Breakfast will be self-service conference breakfast at Eat at the Square

Lunch vouchers can be used in Eat at the Square

Food preferences and intolerances notified on registration will be catered for and clearly labelled.

Location: EPSRC Centres for Doctoral Training Suite, South Kensington Campus, the ICSM Building, Level 4N, Room 402

Speaker: Professor Sofia Olhede

Title: Modeling Network Populations via Graph Distances

Abstract: Networks are a key tool to characterise dependence between nodes or actors. Modern observations often take the form of many networks. In this talk, the core idea is to parametrize a distribution on labelled graphs in terms of a Frechét mean graph (which depends on a user-specified choice of metric or graph distance) and a parameter that controls the concentration of this distribution about its mean. Entropy is the natural parameter for such control, varying from a point mass concentrated on the Frechét mean itself to a uniform distribution over all graphs on a given vertex set. We provide a hierarchical Bayesian approach for exploiting this construction, along with straightforward strategies for sampling from the resultant posterior distribution. We conclude by demonstrating the efficacy of our approach via simulation studies and a multiple-network data analysis example drawn from systems biology.

This is joint work with Simon Lunagomez and Patrick Wolfe

MPE CDT MRes Student Award  – Ben Ashby and Louis Sharrock

MPE CDT MRes Project Award: Elena Saggioro and Niraj Agarwal

MPE CDT Teaching Award – Jennifer Israelsson

MPE CDT Outreach Ambassador Award – Mariana Clare

MPE CDT Inter-Institutional Engagement Award – Adriaan Hilbers

MPE CDT Recognition of research with high societal and environmental impact award – Rhys Thompson

MPE CDT Recognition of research connecting pure and applied mathematics – Stuart Patching

MPE CDT Recognition of interdisciplinary collaboration – Leonardo Ripoli

MPE CDT Oceans Physics Award – Sebastiano Roncoroni

MPE CDT Statistical Mechanics Award – Santos Gutierrez

MPE CDT Dynamical Systems Award – George Chappelle

MPE CDT Numerical Analysis Award – Ieva Dauzickaite

MPE CDT Computer Science Award – Georgios Sialounas

MPECDT Complex Systems Award- Alex Alecio.

MPE CDT Best MRes oral presentations – Swinda Falkena

MPE CDT Best poster presentation- Imogen Dell

Announcing the 5th Annual MPE CDT Jamboree (2019), taking place on 18- 20 March 2019 at the Imperial College London.

The Jamboree provides an excellent forum for multidisciplinary interaction and engagement between industry and academia, as well as the opportunity to exchange ideas and approaches to tackling global challenges in oceans, weather and climate. It also aims to help identify strategies to create innovative adaptation, mitigation and resilience responses to climate change and, importantly, offers the MPE CDT students an opportunity to showcase their research and its impact.

The Jamboree represents the ethos of the MPE CDT by promoting multidisciplinary collaboration, innovation, knowledge sharing and by providing solutions to real-world problems. They will challenge our students to use their skills, knowledge and creativity to work together in collaborative and interdisciplinary teams to propose novel solutions to real world industrial problems.

The final day will bring together a number of eminent researchers, who will give presentations on their work. In addition, our MRes students will deliver blitz talks on the projects they will be working for the rest of this academic year and our PhD students will present posters on their current research work. We would especially like to thank our invited speakers for contributing to and taking part in this year’s MPE CDT Jamboree.

Dr Colin Cotter (Acting Co-Director of the MPE CDT) and Professors Jennifer Scott (Co-Director of the MPE CDT).

See here for the full programme.

Date and Time: Wednesday, 6 March⋅12:00 – 1:00pm

Location: EPSRC Centres for Doctoral Training Suite, South Kensington Campus, the ICSM Building, Level 4N, Room 402

Speaker: Prof. Kody Law

Title: Inference with Multilevel Monte Carlo

Abstract: This talk will concern the problem of inference when the posterior involves continuous models which require approximation before inference can be performed. Typically one cannot sample from the posterior distribution directly, but can at best only evaluate it, up to a normalizing constant. Therefore one must resort to computationally-intensive inference algorithms in order to construct estimators. These algorithms are typically of Monte Carlo type, and include for example Markov chain Monte Carlo, importance samplers, and sequential Monte Carlo samplers. The multilevel Monte Carlo method provides a way of optimally balancing discretization and sampling error on a hierarchy of approximation levels, such that cost is optimized. Recently this method has been applied to computationally intensive inference. This non-trivial task can be achieved in a variety of ways. This talk will review 3 primary strategies which have been successfully employed to achieve optimal (or canonical) convergence rates – in other words faster convergence than i.i.d. sampling at the finest discretization level. Some of the specific resulting algorithms, and applications, will also be presented.

Date : 1 st – 5th July 2019

Location: 402, 4th Floor, ICSM Building, South Kensington Campus, Exhibition Rd, Knightsbridge, London, SW7 2AZ

In this workshop we address the challenge of combining stochastic data-driven mathematical modelling with effective data assimilation techniques in high-dimensional operational systems. The aim of the event is to expose cutting-edge results in stochastic analysis and modelling with nonlinear filtering and data assimilation for applications in geophysical fluid dynamics. We seek to foster collaborations which could eventually bridge the gap between fundamental research in stochastic PDE analysis and data-driven operational methodologies.

Scientific Committee : Colin Cotter, Dan Crisan, Darryl Holm, Wei Pan, Igor Shevchenko..

Further details will be announced in due course.

Registration : please register here by 16th of June 2019

Organisers : Tom Bendall, Erwin Luesink, Oana Lang.

The workshop is funded by the EPSRC Standard Grant EP/N023781/1.

https://www.youtube.com/watch?v=JigJPsyJ0aQ&feature=youtu.be

Applications are invited from PhD students to attend a one-week course on “Multiscale analysis of atmosphere-ocean flows and related numerical issues”.

The course, which will be held at the Met Office in Exeter, is being run by the MPE CDT and the Met Office. The course will be attended by the MPE CDT students who are at the end of the third year of their 4-year programme. In addition, we have a number of places available for external students for which we are now seeking applications.

The principal lecturer will be Professor Rupert Klein from the Freie Universität Berlin

Professor Klein will give five lectures incorporating the following topics:
– Scales in geophysical flows
– Asymptotic techniques for multiscale problems
– Asymptotic analysis of large-scale near-equatorial flows
– Well-balanced schemes and time integration of multiscale systems
– Rigorous justification of reduced asymptotic models

There will also be invited lectures by the following eminent speakers:
– Hilary Weller, University of Reading (numerics for the shallow-water equations)
– Nigel Wood, Met Office (modern NWP model development)
– Beth Wingate, University of Exeter (nonlinear resonances and mean flows)
– Bin Cheng, University of Surrey (analysis of nonlinear dynamics with three time scales)

In the afternoons, there will be supervised hands-on lab sessions, allowing students to perform numerical exercises to illustrate the material in the lectures. Students will need to bring their own laptop with Python 3, numpy and matplotlib pre-installed, and should have experience solving a hyperbolic PDE in one or more dimensions.

The number of places available to non MPE CDT students is strictly limited to 10. These will be allocated to eligible students on a first come, first served basis. The closing date for applications is 10 June 2019.

There is no fee to attend but all students will be responsible for funding and booking their own travel and accommodation.

To apply, please send an email to Sam Williams (s.r.williams@reading.ac.uk) giving brief details of your PhD topic together with a statement from you and your supervisor explaining your suitability for this summer school and how you would benefit.

https://www.hko.gov.hk/hkonews/R2/sidelight_20190208.htm

Location: EPSRC Centres for Doctoral Training Suite, South Kensington Campus, the ICSM Building, Level 4N, Room 402

Speaker: Aretha Teckentrup (University of Edinburgh)

Title Surrogate models in Bayesian inverse problems

Abstract: We are interested in the inverse problem of estimating unknown parameters in a mathematical model from observed data. We follow the Bayesian approach, in which the solution to the inverse problem is the probability distribution of the unknown parameters conditioned on the observed data, the so-called posterior distribution. We are particularly interested in the case where the mathematical model is non-linear and expensive to simulate, for example given by a partial differential equation. We consider the use of surrogate models to approximate the Bayesian posterior distribution. We present a general framework for the analysis of the error introduced in the posterior distribution.