Floods precipitate weather network upgrade

26 September 2014

The Met Office’s current network consists of 15 operational weather radars across the UK such as the station at Great Dun Fell in Cumbria.

Following last winter’s floods, the Met Office is halfway through an upgrade to the network that links its 15 weather radar stations to its Exeter HQ. The upgrade, which includes advanced weather radar and software, aims to improve forecasting for the next 15 to 20 years.

The Met Office Remote Site Network is a private WAN and has now been outsourced to BT which is upgrading the links to each site to 2Mbps symmetric connections. The circuits are provided by Openreach’s MPLS infrastructure and managed as part of BT’s Convergent Solutions. They use the firm’s Connect IP Clear system which provides symmetric bandwidth up and downstream. 

The remoteness of the sites and the need for 2Mbps circuits meant the telco had to replace copper cable with fibre or microwave. The Met Office says its radar sites are less suited to ADSL (asymmetric) data transfer as these circuits provide a lower and more restricted upstream data path. 

It adds that the new 2Mbps infrastructure provides sufficient bandwidth for the radar data transfer and a “reasonable” data rate with its new enhanced product data set.

The Met Office’s current network consists of 15 operational weather radars across the UK such as the station at Great Dun Fell in Cumbria. These provide real-time information to help monitor and forecast heavy rainfall to issue flood warnings. 

Many of the transmitters were more than 20 years old and suffered recurring age-related failures. The new dual polarisation radar and new Cyclops software has given greater reliability and allows more flexibility for radar R&D activities in the future. 

Cyclops is said to provide better filtering of raw radar data to remove “ground clutter” (echoes) on precipitation rates and accumulation (flooding). The Met Office says this makes it easier to distinguish precipitation and non-precipitation echoes, identify precipitation type, better correct for attenuation of the radar signal due to very heavy rain or hail, and estimate drop shape. These in turn will improve precipitation estimation.