On a cold night in winter with a frost forecast, UK highway authorities potentially spread up to 35,000 tonnes of rock salt over 3,500 salting routes in order to keep the nation’s key roads open and safe.  The spreading of de-icing agents in such quantities is not only a financial burden in terms of the costs of applying the salt and the damage it causes to concrete structures, but it can also be very damaging to the environment.  To help with daily decision making (i.e. to salt or not), highway engineers consult a forecast as part of a Road Weather Information System.  However, until recently, there has been a paucity of high resolution observations meaning that forecasts cannot be verified / improved significantly away from dedicated outstations on the road network.  On a typical night in winter, there can be as much as 10-15°C variations around the road network, which means if forecasts can be improved, then gritters can be fine-tuned to selectively salt only the coldest sections, saving considerable money.

The Internet of Things is changing this.  The University of Birmingham has invented the wintersense product which is a self-contained low cost road surface temperature sensor.  It uses a lithium battery (2 year lifetime) and Low Power Wireless Area Networks (LPWAN: Internet of Things) communications to produce a consumable priced solution to road surface temperature sensing.  Each unit costs £hundreds as opposed to the £thousands typically spent by highway authorities on current outstation solutions.  Unsurprisingly, given this large difference in price, there has been significant interest from the sector and networks are now being deployed across the UK.  This studentship will look at how this new high resolution data can be better utilised in forecasting applications to maximise the benefits from the sector. 

Wintersensor deployed on the University of Birmingham Campus


This studentship will extend the current IoT sensing approach to include a nowcasting solution.  Nowcasting (using a combination of statistical and numerical modelling techniques) is essentially short range forecasting which takes current observations to estimate the weather for a short period of time into the future.  Therefore, by closely monitoring temperature trends obtained at the sensors, along with more detailed weather information from outstations, a 3 hour local forecast could be produced to provide sufficient warning to prepare the fleet and secure the network before ice forms.  For marginal nights, the information from the sensors could be closely watched, deploying the gritters only when absolutely necessary and thus making considerable savings.  The nowcasts could then be embedded within existing forecast solutions to provide a comprehensive new approach.

Training and Skills

CENTA students are required to complete 45 days training throughout their PhD including a 10 day placement. In the first year, students will be trained as a single cohort on environmental science, research methods and core skills. Throughout the PhD, training will progress from core skills sets to master classes specific to the student's projects and themes. 

Additional training will be provided to the successful candidate in the role of the Internet of Things in environmental monitoring.  It is envisaged that the student will this become experienced in designing and deploying sensor networks as well as producing server side data solutions and modelling capability.  Hence, the project will suit a numerate graduate with an interest in applied meteorology.


Year 1: Exposure to Sensor Network design and deployment for winter road maintenance.  Develop understanding of the sector and explore current and new options for forecasting.

Year 2: Development of nowcasting forecast modules using data collected over the last 3 seasons.  Presentation at the Standing International Road Weather Conference.

Year 3: Test models in an operational setting using real-time data.

Partners and collaboration (including CASE)

The successful candidate will work closely with a new University operating division called altasense.  The fledgling spin-out company focuses on reducing the weather sensitivity of transport infrastructure by using low cost sensing techniques.  So far, it has primarily explored the rail market (leaves on the line) and the road market (winter road maintenance).  An additional CASE partner from the weather forecasting industry will also be involved (tbc), providing additional training in this particular area.  A local authority partner for the trial will also be recruited (existing altasense customer)

Further Details

Contact Prof. Lee Chapman: l.chapman@bham.ac.uk

See also the following website: