Alarms for Church Roofs
Angus Brown
|
|
Stoke Minster, Stoke-on-Trent: when caught, the perpetrators
were found to have been paid £480 for the lead they had taken
from the roof, but the damage they had caused
cost more than £33,000 to repair. |
|
Although lead theft is not such a
regular news item as it was in 2013, the
threat to historic buildings from this
crime has not gone away. Lead theft is directly
linked to its scrap value which has soared
during the last six years from less than £400
per tonne to a peak of more than £1,200. With
a current price of £950 per tonne, churches are
still being targeted for the lead and copper on
their roofs.
For example, between 2000 and
2004 there were just 20 thefts of metal a year
from churches across the country, however
between 2007 and 2011 these increased to
more than 14,000 reported cases, at a cost of
more than £32 million.
Churches are perceived
as fairly easy targets, particularly in isolated
settings, and specialist insurer Ecclesiastical
(EIG plc) reports that it is still receiving
more than ten metal theft claims per week.
The true cost of stripping lead from a
church roof is, of course, not simply the cost
of the replacement material: there is often
a very wide discrepancy between the small
quantity of metal stolen and the damage
recklessly caused in the execution of the theft.
Metal thieves who target historic churches
cause damage which goes far beyond the
value of the metal that is stolen. In some
cases, the damage caused by rainwater ingress
extends to medieval wall paintings and other
irreplaceable pieces of art and historic fabric.
There is clearly a graduating scale of metal
thieves which ranges from the ‘chancer’, the
opportunistic and small-scale offender who is
after quick cash, to the highly organised and
specialist criminal who will strip entire roofs.
In general, however, research has shown that
the sophistication of thieves targeting metal
roofs tends towards the lower end of the scale.
There has also been concern regarding
scrap metal dealers, with their ‘no-questions-asked’
approach and an industry-wide refusal
to ban cash-based transactions. Only when
the government introduced a new law in
England and Wales regarding these issues were
dealers south of the border forced to adapt.
DETERRENCE
|
|
|
The first line of defence: padlocked gates at Seend, Wiltshire display notices advising of security precautions
in use, which include roof alarms. |
Since lead thieves who are interrupted can
still cause expensive damage, the use of 24-hour
monitored alarms with audio or CCTV
verification is increasingly considered to be
ineffective in protecting the historic fabric of
an ancient church. The potential thieves need
to be deterred before they do any damage.
Deterrence starts at the perimeter, with
notices advising would-be burglars of the
measures that have been taken to prevent
theft. These may include the use of physical
deterrence systems such as alarms, anti-climb
paint, ‘DNA’ marking systems which enable
stolen material to be identified, and lead sheet locking systems that make it difficult
to strip a roof of its lead quickly and quietly.
Deterrence may also include community involvement
through neighbourhood watch schemes or
notices inviting members of the
public to call the police if they see suspicious
vans or workmen between 6pm and 8am.
Visibility is another key component of
this first layer of defence. Is the building
overlooked by houses, does any vegetation need
to be cleared to provide a better view? Most
criminals can be deterred by knowledge that
the site is regularly overlooked and checked.
The second layer in the deterrence
approach is an alarm system. Its aim is to ensure
that intruders are detected and deterred the
instant they go where they should not be and
crucially, before they have done any damage.
ACCESSIBILITY
When looking at the security of a historic
church it is essential to remember that the
building is regularly open to the public and
although it may contain important historic
fabric and have a lead roof, the public needs to
be able to access the site without impediment.
There are many ways of impeding access
by would-be thieves and these can be as
simple as making sure that ladders and tools
are secured and not accessible. Looking at
the perimeter of the site, can it be ensured
that only authorised vehicles have access?
Looking at the building itself, are there
parts of the building design which might
enable easy access to the roof? A boiler house might provide the perfect spot to climb up
onto the aisle. In these cases are there any
other physical measures that can be installed
to help deter access to the roof, such as anti-climb
paint or security lighting. The key issue
for historic churches is to ensure that any
physical intervention does not compromise
the appearance of the building, so 18ft-high
razor-wire fences are not always possible.
|
|
|
Left: damaged plasterwork at All Saints, Newton Heath,
Manchester following the theft of its lead. Centre: tucked out of site from the surrounding houses,
areas of lead roof like this, on a church in the diocese
of Durham, are easy prey without an alarm. Right: a control panel and sounder in a church tower (as all
the detectors are wireless and battery powered, this is
the only cabling required). |
STRATEGY
A truly comprehensive security management
plan blends physical, electronic and procedural
aspects of protection, while respecting the
context in which it is being installed. So not only
do you look at the simple actions mentioned
above but also the protection that technology
can provide and how those people who are
responsible for the building behave.
This is
fundamental to the process of developing a
comprehensive security strategy and without
internal policies and procedures – as well as
a proactive security consciousness – physical
and electronic security hardware and
features will fall short of their intended goal.
A security strategy is developed by evaluating
the assets, considering the threats against
them and developing countermeasures to
reduce their vulnerabilities. This is achieved
by layering security, both physical and
electronic, thereby creating defence in depth.
This approach to security is designed to
deter a crime in the first instance, to delay access
to and the removal of target assets and to detect
an attack at an early stage, should one occur.
Thus potential criminals, having inspected the
property, may be deterred from undertaking a
burglary by the sight of robust physical security,
intrusion sensing and external lighting. Should
they decide to proceed, physical measures
will delay access to target assets and the alarm
system will provide an early indication of attack
and may summon an effective response.
THE ALARM SYSTEM
In 2010 alarm specialists E-Bound AVX
Ltd worked with Ecclesiastical Insurance
Group (EIG plc) to refine a simple but
effective strategy for the protection of
historic church buildings as part of EIG’s
‘Hands off our church roofs’ campaign. The
three key requirements identified were:
- Wireless: this ensures that there are no
cables and fixings marring the building,
and avoids the need to drill holes in historic
structures (including walls often over a
metre thick).
- Robust: this means that the alarm will trigger when it should and not when it
should not. It is essential that the product
is reliable and is capable of distinguishing
criminal threat from background activity.
- Simple: the alarm system should be
as simple to operate as possible as the
majority of churches are cared for by
volunteers and responsibility for the alarm
can change frequently. No church wants
a lengthy instruction manual or intense
training programme, they just want the
alarm to work.
|
|
Little more than 20 detectors were needed to protect
the roof at Tewkesbury Abbey: these two form part of
the protection for the ambulatory. |
|
|
|
A wireless movement detector at St Mary’s, Yelden,
Bedfordshire: for church roofs, movement detectors
often require bespoke lenses and sensors, in this case
background movement from the churchyard below
must be eliminated so that it responds to movement
at parapet level only. |
|
|
|
One of the two sounders required at Sheffield
Cathedral. The one shown here with a strobe next to
it, is above the roof of the sacristy and focusses the
sound onto the roof below. |
|
A typical system consists of a control
panel, wireless passive infrared movement
detectors, audio alarm sounders and strobe
lighting. The detectors are battery operated
and communicate with the control panel
via radio signal using a frequency designed
to suit the thick walls, lead roofs and
other features typically found in church
architecture. As a result, the only cabling
required is from the control panel to the
mains power supply and to the sounders,
and a typical installation can usually be
deployed in one day, subject to survey.
The control panel may be installed in the
church or an ancillary building and should
be connected by mobile phone technology
to a 24-hour monitoring and response
centre. On the roof of the church, the audio
alarm typically involves a short burst of a
siren followed by a pre-recorded challenge
warning the intruder to leave immediately
and making clear that people are responding
to the alarm. The message ends with another
short burst of the siren. A sequence like this
which includes a good, clear spoken challenge
has been shown to be far more effective at
deterring intruders than a siren alone.
Despite many similarities, each historic
place of worship is unique, and each alarm
system must be designed on a bespoke basis.
There is no off-the-shelf solution: for the alarm
system to work effectively, each site needs to be
considered on its own merits, and each detector
needs to be tailored to suit the parameters
of its location. Not only are there variations
in design and construction to be taken into
consideration, but there are also variations
in the requirements of the people, usually
volunteers, who look after the building. Both
sets of requirements must always be considered
carefully for an alarm system to be effective.
Where historic fabric is concerned,
it is also important that the alarm system
has been developed in consultation with
historic building professionals, to ensure
that the system is as compatible as possible
with historic fabric. Equipment should be
as small and unobtrusive as possible and
great care must be taken over the fixings.
The positioning of detectors and panels must
be thought through for ease of installation
and operation, as well as for aesthetics.
The alarm system must be based on the
proven catch capabilities of industry-standard
detectors but without the inherent risk of
regular false alarms usually associated with
outdoor installations. The best roof alarm
systems have been developed as roof alarms
from the outset. Adaptations of existing
systems, designed for homes, factories, shops and the like will not be able to cope with
the unique challenges posed by buildings
with extremely thick walls and a variety of
no-go areas which must be monitored in isolation from other areas. Detectors often
need to be tailored to suit odd angles, or to
eliminate background radiation or movement,
requiring a range of lenses and sensors.
Monitoring is essential so that there is
a response when the alarms are set off. In
addition, systems should include tamper
protection on all detectors and the control
panel, along with mains power monitoring.
The company monitoring the alarm should
also be alerted in the event that mains power
is lost and not restored. The alarm should have
rechargeable back-up batteries which will run
the system for several days in the event that
mains power is lost. Once power is restored,
these batteries automatically re-charge.
When integrated into a broad strategy
for the protection of churches, dedicated roof
alarm systems provide a simple but highly
effective form of deterrence.
At Puddletown
in Dorset, for example, St Mary’s Church
had a spate of attacks in mid-summer 2010,
striking at the lead valley between the nave
and aisles. What was particularly galling
was the fact that the thieves cut out the base
of the gutter near the outlet, so water from
the whole expanse of roof came through
the hole into the church below, causing
extensive damage. It was also very difficult to
make a temporary repair watertight in this
area.
An alarm was installed in September
2010. Within two weeks of the work being
completed the PCC was advised that the alarm
had been set off at 2.30am on a Saturday.
On attending the church that morning,
finger marks were found in the anti-climb
paint on a drain pipe and there was evidence
that an attempt at a further theft had been
made, this time without causing any further
damage. Now, four years on, the installation
is still working well and has attracted the
attention of Dorset Police crime reduction
officers and other churches in the region.
~~~
Further Information
ChurchCare (the Church Buildings Council’s
website) publishes various guidance notes including
one on alternatives to lead
Ecclesiastical (EIG plc) provides information on
lead theft protection
English Heritage co-ordinates the Alliance to
Reduce Crime against Heritage (ARCH),
a voluntary national network which
includes churches, police authorities, local
authorities and other organisations
Historic Scotland publishes a short guide on
lead theft and traditional buildings
|