Please find details below of an available PhD project at Southampton. Informal enquiries via Dr Sujit Sahu (as below).
----------------------------
Forecasting Air Pollution through Novel Space-Time Modelling Methods
University of Southampton - School of Mathematics and Southampton Statistical Sciences Research Institute
Ground-level ozone is an air pollutant that is a significant health
risk, especially for children with asthma. It also damages crops, trees
and other vegetation and is a main ingredient of urban smog. To
evaluate exposure and forecast ozone levels, the United States
Environmental Protection Agency (USEPA) has developed several sparse
network of monitoring stations covering the whole of the United States
(US). Data obtained from these sparse networks must be processed using
stochastic spatial and spatio-temporal models to make valid inference
for ozone levels at particular sites, such as urban areas, based on
rigorous statistical methods.
The overall aim of this project is to develop new stochastic methods of
forecasting air pollution levels, together with software necessary for
their implementation. These methods may supersede the non-stochastic,
site-specific regression methods currently used by many government
agencies. The research will impact on forecasting various different
kinds of air pollutants in any geographical region.
The first objective of this project is to develop improved forecasting
methods together with estimates of uncertainty for the `eight-hour
average map'. This map of ozone levels is the average for the past four
hours, current hour and three hours ahead of the current hour. The
estimates of uncertainty are vital for assessing accuracy and making
informed decisions. The forecasting methods will combine novel
space-time weighting of real-time air monitoring data and forecasts
from a deterministic computer simulation model known as the Community
Multi-scale Air Quality model. The second objective of this project is
to develop further methods for forecasting the daily maximum eight-hour
average map for the whole of the US. Validation of the methods will be
performed using data from monitoring stations not used for estimation
and forecasting purposes.
The project is part of an on-going collaboration between S3RI and the
USEPA. The USEPA will contribute in-kind by supplying the necessary
data and providing expert staff time to evaluate the methods developed
and software. There is also the possibility for the student to visit the
USEPA to enhance practical understanding of the project.
The successful applicant will have a good BSc or MSc degree in
Mathematics, Statistics or a closely related field.
The closing date for applications is May 3, 2011.
Please address enquiries to Dr Sujit Sahu ([log in to unmask]). For more information on Statistics at Southampton, please visit http://www.southampton.ac.uk/s3ri and in particular two shortcourses currently organised by Dr Sahu on Bayesian and Spatial Statistics (http://www.s3ri.soton.ac.uk/courses/hierarchicalmodelling/index.php) scheduled for June 7-10, 2011 may also be of interest.
An online application form is available: http://www.soton.ac.uk/postgraduate/pgstudy/howdoiapplypg.html (in the form select: School of Mathematics, research), together with general information on postgraduate study at the University.
You may leave the list at any time by sending the command
SIGNOFF allstat
to [log in to unmask], leaving the subject line blank.
|