Guide to the use of EN 1991-1-4 – Wind Actions

Peak velocity pressure qp(z) (4.5(1)): is the site wind velocity taking account of the terrain and building
size. qp(z) corresponds to qs in BS6399-2.

Probability factor cprob (4.2(2)P Note 4): used to modify the basic wind velocity to change the risk of the
wind speed being exceeded. cprob is the same as Sp in BS6399-2.

Resonant response factor R (6.3.1(1): accounts for the effects of wind turbulence in resonance with the
vibration of the structure in its fundamental mode of vibration. There is no equivalent value in BS6399-2.

Roughness factor cr(z) (4.3.1(1)): used to modify the mean wind speed to account for the terrain roughness
upwind of the site and the height of the building or structure under consideration. cr(z) corresponds to Sc in
the BS6399-2 directional method.

Season factor cseason (4.2(2)P): used to modify the basic wind velocity to produce wind speeds with the
same risk of being exceeded in any specific sub-annual period. cseason is the same as Ss in BS6399-2.

Structural factor cscd (5.3(2)): takes account of the effect of non-simultaneous wind action over the
surfaces of the structure or element combined with the effect of dynamic response of the structure or
element. cs and cd correspond to Ca and (1 +Cr) respectively in BS6399-2.

Turbulence intensity Iv(z) (4.4(1)): is the standard deviation of the wind turbulence divided by the mean
wind velocity and is a measure of the gustiness of the wind. Iv(z) corresponds to StTt in the BS6399-2
directional method.

Wind force Fw , Fw,e , Fw,I , Ffr (5.3(2): the wind force acting on the overall structure or element (Fw), on the
external surfaces (Fw,e), on the internal surfaces (Fw,i) or due to frictional forces (Ffr). Corresponds to P and Pf
in BS6399-2.

Wind pressure we and wi (5.2(1)): the wind pressure acting on external and internal building surfaces
respectively. Corresponds to pe and pi in BS6399-2.

2.2 Scope of EN1991-1-4

EN1991-1-4 is applicable to:

• Building and civil engineering works with heights up to 200m

• Bridges with spans of not more than 200m (subject to certain limitations based on dynamic
response criteria)

• Land based structures, their components and appendages

The specific exclusions are:

• Lattice towers with non-parallel chords

• Guyed masts and guyed chimneys

• Cable supported bridges

• Bridge deck vibration from transverse wind turbulence

• Torsional vibrations of buildings

• Modes of vibration higher than the fundamental mode

The scope of EN1991-1-4 is much wider than BS6399-2, it includes wind actions on other structures, which
in the UK are given in a number of other British Standards and design guides. In some cases, for example

8
 EN 1991: Actions on structures – Part 1.4: Wind Actions

dynamic response of buildings, there is no equivalent UK standard. Table 2 shows the scope of EN1991-1-4
compared with current UK practice, (where no direct equivalent UK current standard is available, the nearest
other guidance is shown in brackets).

Table 2 Comparison between EN1991-1-4 and current UK practice

EN1991-1-4 UK Practice

Buildings (static) BS6399-2

Buildings (mildly dynamic) BS6399-2

Buildings (dynamic) No direct equivalent (ESDU)

Bridges Design manuals for roads and bridges, BD49, BD37

Chimneys (Steel) BS4076, CICIND

Chimneys (concrete) CICIND (Int. Committee for Industrial Chimneys)

Scaffolding BS5975, BS5973, BS EN 12811

Structural elements Limited guidance in BS6399-2 (CP3-V-2)

Lattice structures (not towers) No direct equivalent (CP3-V-2)

Spheres, domes, barrel vaults No direct equivalent (reference books, papers)

2.3 UK National Annex EN1991-1-4

EN1991-1-4 must be used in the UK with the accompanying National Annex. The National Annex (NA)
contains information on National Determined Parameters (NDPs) such as:

• Values and/or classes where alternatives are given in the Eurocode

• Values to be used where only a symbol is given in the Eurocode

• Country specific geographical or climatic data

• Procedures to be used where alternative procedures are given in the Eurocode

• Decisions on the application of informative annexes

• References to non-contradictory complementary information to assist the user

A National Annex cannot change or modify the content of the EN text in any way other than where allowed
for above.

In EN1991-1-4 there are 61 NDPs; 51 in the main body and 10 in the annexes. In most cases a
recommended procedure or value is given, although national choice is allowed. The UK National Annex
provides an alternative for a number of these procedures. These are listed in Table 3 with a commentary on
the reasons for the alternative procedures (additional alternative procedures are also given in Section 8 Wind
actions on bridges and in the annexes, but these are outside of the scope of this report).

9
Guide to the use of EN 1991-1-4 – Wind Actions

Table 3 Clauses where the UK National Annex gives alternative rules or guidance
EN1991-1-4 Commentary
Clause number

4.1(1) Peak velocity pressure qp - Procedures for directly obtaining qp are given in 4.5
4.2(1)P Note 2 Fundamental value of the basic wind velocity vb,0 — this is the wind velocity at a height of
10m above ground level. The UK wind velocity map is for a height of 10m above sea level
(Figure NA1); an expression is given to adjust the UK map velocity using the altitude factor.
4.2(2)P Note 1 Altitude factor calt — an expression is given forcalt for use in the UK. Note that this is not the
same as the factor used in BS6399-2, it now allows a reduction in calt with height above ground
4.2(2)P Note 2 Directional factor cdir - an expression is given for cdir for use in the UK
4.2(2)P Note 3 Season factor cseason - an expression is given for cseason for use in the UK
4.3.2(1) Note 1 Roughness factor cr(z) - Expressions 4.4 and 4.5 are replaced by two figures, NA3 and NA4,
which directly give cr(z). These figures account for the upwind distance to sea and for urban
sites, the distance to the edge of the urban area. The opportunity has also been taken to
simplify the five terrain categories to Sea, Country and Town.
4.4(1) Note 2 Turbulence factor ki/ln((z-hdis)/zo) — a new turbulence factor has been introduced to account
for up wind distances to sea and edge of town, this is given in Figures NA5 and NA6.
4.5(1) Note 1 Peak velocity pressure qp - Expression 4.8 has been replaced by expression NA3. This was
necessary because 4.8 is a linearised form of the equation for qp which underestimates wind
pressures, particularly in Town terrain. NA3 corresponds with the approach used in BS6399-2.
A new term, the exposure factor for flat terrain ce,flat(z), has been introduced in Figures NA7 and
NA8. This greatly simplifies the determination of qp(z) for sites on flat terrain or on hilly sites at
heights above ground of up to z= 50m.
Air density ρ — The value for air density in the UK is 1.226kg/m (as used in BS6399-2)
2
4.5(1) Note 2
5.3(5) Lack of correlation between windward and leeward faces - the recommended procedure
only allows the reduction factor in 7.2.2(3) to be applied to walls. In the UK, the reduction will be
applied to all horizontal components of the wind force acting on walls and roofs. This will
reduce overall wind loads and is similar to the approach used in BS6399-2.
6.6(1) Structural factor cscd — In the UKcscd may be separated in to a size factor, cs, and a dynamic
factor, cd, using Table NA3 and Figure NA9. This simple approach avoids the need to use the
complex dynamic procedures in 6.3 and can give significant reductions in wind loads on large
buildings or elements that are not dynamically sensitive. The cs and cd factors correspond to the
size effect factor, Ca, and dynamic augmentation factor ,Cr, in BS6399-2.
7.1.2(2) Asymmetry — No decision has yet been made for the UK National Annex
7.2.1(1) Note 2 Determining the external pressure coefficient cpe — the recommended procedure requires
2 2
pressure coefficients to be calculated for every loaded area between 1m and 10m using
logarithmic interpolation. This would increase the calculation effort significantly in the UK
cladding and glazing industry. There is no scientific evidence to support this rule, therefore in
2
the UK, the cpe,1 value should be used for all loaded areas † 1m and the cpe,10 value for areas
2
>1m . This new procedure was developed in consultation with the UK glazing industry.
7.2.2(2) Note 1 External pressure coefficients for walls - Table 7.1 is replaced by Table NA4. This was done
to allow the inclusion of net pressure coefficients, which can give significant reductions in overall
wind loads. This corresponds with the new Table 5a in BS6399-2.
7.2.8(1) Vaulted roofs - Figure 7.11 is replaced by Figures NA10 and NA11. These new figures give
better estimates of pressure coefficients on vaulted roofs based on recent wind tunnel studies.
7.2.9(2) Permeability - Table NA5 gives typical permeability values for UK construction
7.2.10(3) Note 2 Multiskin walls and roofs — Experience in the UK has shown that the recommended
procedure does not apply to tiled and slated roofs or masonry cavity walls. The EN1991-1-4
procedure should not be used in the UK for these situations. The existing guidance in BS5534
and BS5628-1 should continue to be used.
7.13(2) Effective slenderness length - Figure 7.16 is replaced by NA6. This is because the values in
Figure 7.16 are of uncertain origin and are significantly different from the values currently used
in the UK. The revised values given in NA6 correspond with those given in BS6399-2

10
 EN 1991: Actions on structures – Part 1.4: Wind Actions

2.4 Main differences between EN1991-1-4 (with the UK NA) and BS6399-2

When considering wind loads on static building structures there are many similarities between EN1991-1-4
and BS6399-2, there are also a number of obvious differences, such as the change from mean hourly to 10-
minute mean wind speeds plus other more subtle differences in the methodologies and expressions used;
some of which will not be immediately obvious to the user. The UK National Annex also gives some
alternative procedures which change EN1991-1-4 to make it applicable for use in the UK. This Chapter
provides an overview of the first seven sections of EN1991-1-4 (excluding section eight on bridges) and
identifies and discusses the major differences from BS6399-2. Table 4 shows some of the main differences
between EN1991-1-4 and BS6399-2. For a detailed commentary on EN1991-1-4, the reader is referred to
Cook [3].

2.4.1 Section 1 General

This section gives the scope, normative references, assumptions, rules for design assisted by testing and
measurements, definitions and symbols. The differences in definitions, symbols and scope have already
been explained in section 1 of this report. Section 1.5 Design assisted by testing and measurements, allows
wind tunnel testing, validated numerical methods or full scale testing to be used to obtain the wind loads and
the structural response. This is essentially the same as the guidance given in BS6399-2 which allows other
equivalent methods to be used. However, whereas BS6399-2 gives guidance for wind tunnel testing in
Annex A, EN1991-1-4 does not give any guidance. It is likely that guidance on this and other parts of
BS6399-2 which contain non conflicting complementary information (NCCI), such as wind effects on irregular
shape buildings, inset stories, multipitch roofs, etc. will be published as a residual standard.

2.4.2 Section 2 Design situations

The general requirement of this section is that the relevant wind actions be determined for each design
situation identified in EN1990 (Basis of Structural Design), such as persistent, transient and accidental
design situations. In EN1991-1-4, the following specific design situations should also be taken in to account:

• other actions (such as snow, traffic or ice) which will affect the wind loads

• changes to the structure during construction which modify the wind loads

• where doors and windows that are assumed to be shut are open under storm conditions then they
should be treated as accidental actions

• fatigue due to wind actions should be considered for susceptible structures

This section does not have a direct equivalent in BS6399-2. The guidance for doors and windows
accidentally left open or which break during a wind storm is different from that given in BS6399-2. In
BS6399-2 this is treated as a serviceability limit state and a reduced probability factor of 0.8 used which has
the effect of reducing the mean recurrence interval from 50 to about 2 years. In EN1991-1-4 this is treated as
an accidental design situation.

2.4.3 Section 3 Modelling of wind actions

This Section describes wind actions and defines characteristic values of wind velocity and velocity pressure
according to EN1990. Wind actions are classified according to EN1990 as variable fixed actions; this might
seem like a contradiction in terms but it essentially means that for a given structure in a given location the
wind action is fixed except where a range of values is given. For example on a pitched roof both positive and
negative pressure coefficients are given for many roof zones, both of which should be considered in design.
The information given in this section is consistent with BS6399-2, although there is no direct equivalent to
this section in BS6399-2. Definitions of fixed and variable actions are given in EN1990.

11
Guide to the use of EN 1991-1-4 – Wind Actions

2.4.4 Section 4 Wind velocity and velocity pressure

This section gives the procedures for determining the wind velocity and the peak velocity pressure.
Alternative procedures have been given in the NA for a number of clauses in this section in order to give a
more realistic representation of wind conditions for the UK, these are described below.

2.4.4.1 Basic wind velocity

The starting point for the determination of the wind velocity is the map of fundamental basic wind velocity
given in Figure NA1 of the NA. This map corresponds to the map of basic wind speed given in BS6399-2,
except that it is based on a 10-minute mean velocity instead of the mean-hourly map used in BS6399-2 and
a slightly increased terrain roughness (z0 = 0.05 instead of 0.03). These changes do not account for the
major differences between these maps. This is caused by a reanalysis of the meteorological wind speed
data which has resulted in slightly higher windspeeds over most of the country except for the Southeast
where quite large reductions are seen.

The map wind speed is adjusted using the altitude factor (given in the NA) which differs from that given in
BS6399-2 because it now reduces with height above ground. This new altitude factor is based on a study
[10] which showed that the old altitude factor in BS6399-2 was very conservative, especially for site altitudes
above 200m.

The basic wind velocity is determined from expression (4.1), which is essentially the same as equation (8) in
BS6399-2.

v b = cdir ⋅ cseason ⋅ v b ,0 (4.1) from EN1991-1-4

Vs = S d ⋅ S s ⋅ S a ⋅ S p ⋅ Vb (8) from BS6399-2

Where vb,0 = Vb = map wind speed (vb,0 is adjusted for altitude in the NA using calt)
cdir = Sd = direction factor (the same procedure is used in both standards)
cseason = Ss = season factor (the same procedure is used in both standards)
calt = Sa = altitude factor (calt uses an improved, less conservative, version of the BS6399-2 procedure)
cprob = Sp = probability factor (the same procedure is used in both standards) note that cprob is not included
in expression 4.1 but is allowed in Note 4 of 4.2 (2)P

2.4.4.2 Mean wind velocity

There is no direct equivalent of the mean wind velocity term in the main body of BS6399-2. However,
multiplying So, the terrain and building factor for hourly mean windspeeds given in Annex C.2.1 by the site
windspeed vs gives the mean wind velocity corresponding to the EN1991-1-4 value. In EN1991-1-4 the mean
windspeed is used extensively in Annexes B, C and E for determining inwind and vortex shedding response.
It is also used to determine the peak velocity pressure as described in 4.4.4 below.

The mean wind velocity vm is given by expression 4.3.

v m ( z ) = c r ( z ) ⋅ co ( z ) ⋅ v b (4.3)

where cr(z) is the roughness factor which accounts for the ground roughness of the upwind terrain

co(z) is the orography factor which accounts for the effects of topography at the site

The procedure for determining the orography factor is given in Annex A.3 of EN1991-1-4 and is the same as
that used in BS6399-2.

The roughness factor assumes an equilibrium wind velocity profile, which implies that the boundary layer is
in equilibrium at all heights and does not change with distance downwind. Such a profile could only be
established with an upwind fetch of approximately 200km of flat uniform terrain; such conditions do not occur
in the UK. Thus in order to more realistically account for changes in wind speed with terrain, the UK NA gives

12
 EN 1991: Actions on structures – Part 1.4: Wind Actions

an alternative procedure for determining cr(z) using Figures NA3 and NA4. Because of this alternative
procedure, expression (4.4) for the roughness factor and (4.5) for the terrain factor are not to be used in the
UK.

Other points worth noting in this Section are:

• The terrain categories - Table 4.1 defines five terrain categories ranging from sea 0, to city centre IV.
For ease of use, the NA reduces the number of terrain categories to three which approximately
correspond to the three terrain categories in BS6399-2. These are Sea (category 0), Country
(categories I and II) and Town (categories III and IV).

• Increased wind speeds from surrounding buildings - Section 4.3.4 gives guidance to account for the
wind speed-up caused by large and considerably taller neighbouring structures. The recommended
procedure, given in Annex A.4, is simple to use and generally conservative. This should prove to be
more helpful to the user than the guidance given in BS6399-2 which just gives a warning and
suggests that specialist advice should be sought.

• Shelter effects - Section 4.3.5 allows the effects of shelter from closely spaced surrounding buildings
to be taken in to account. The recommended procedure, given in Annex A.5, is the same as the
procedure used in BS6399-2 and uses the displacement height hdis. However, EN1991-1-4 does not
give any guidance on how hdis should be used. In general the building height z may be replaced with
z – hdis. The new figures for the roughness factor, Figures NA3 and NA4, are both given in terms of z
– hdis.

2.4.4.3 Wind turbulence

The turbulence intensity given in EN1991-1-4 by expression (4.7) is for equilibrium conditions, which as
discussed above do not occur in the UK. The effect of distance to sea and distance to edge of town may be
accounted for by the turbulence factor, ki, which has a recommended value of 1.0. In the UK this
recommended value is not used. Instead, the NA gives an alternative procedure for determining ki.
However, to make calculation easier graphs of the factor ki/ln((z-hdis)/zo) are given, which for terrain where
orography is not significant gives the turbulence intensity directly. This factor is given in Figure NA5 for sites
in Country terrain with a correction factor for sites in Town terrain given in Figure NA6.

2.4.4.4 Peak velocity pressure

EN1991-1-4 uses a simplified expression for the peak velocity pressure given in expression 4.8

qp = 1/2 ρ vm [1 + 7Iv(z)] = ce(z) qb

2
(4.8)
2
Where ce(z) is the exposure factor which corresponds to Sb in BS6399-2; except that in EN1991-1-4 a
linearised version is used which disregards second order turbulence effects and consequently can
significantly underestimate the wind pressure, particularly in urban terrain where the turbulence is greatest.
This additional turbulence term is included in the BS6399-2 method (see box).

For this reason expression 4.8 in EN1991-1-4 is not used in the UK and has been replaced by expression
NA3

qp = 1/2 ρ vm [1 + 3Iv(z)] = ce(z) qb

2 2
(NA3)

To simplify the calculation of qp, the NA introduces a new term, ce,flat which is the exposure factor for sites
where orography is not significant. ce,flat is obtained directly from Figures NA7 and NA8 and replaces ce(z) in
expression NA3. When orography is significant but the height of the structure is less than 50m then the NA
gives a simplified conservative procedure for determining qp using expression NA4.

13
Guide to the use of EN 1991-1-4 – Wind Actions

Comparison between ce(z) and Sb2

2
In EN1991-1-4 the exposure factor ce(z) corresponds to the terrain and building factor Sb in BS6399-2
2 2 2
From BS86399-2 equation 28, Sb = Sc (1 + (gt St) + Sh) (for Country terrain)
Making the substitution S0 = (1 + Sh) gives
2
2  g S 
Sb = Sc So 1 + ( t t
2 2
)
 So 
  2g S   g t St 
2
S b = Sc S o
2 2 2
1 +  t t +  
  So   S
  o

 
 

2 2
Compare this with expression 4.8 in EN1991-1-4, ce(z) = cr (z) co (z) [( 1 + 7Iv(z)] , where 7Iv(z)
2 2 2 2
corresponds with 2gt St/So and cr (z) co (z) corresponds with Sc So . It can be seen that there is no term
2
equivalent to (gt St/So) . The effect of this missing term is to underestimate the wind pressure. In the worst
case of a very low building in category V terrain this could be by up to 30% (depending on the turbulence
intensity).

qp = [qb ce,flat(z)] [(co + 0.6)/1.6] (NA4)

For structures higher than 50m then expression NA3 should be used.

2.4.4.5 Value for air density

The recommended value for air density, ρ, in EN1991-1-4 is 1.25kg/m . The UK has historically used a value
3

3
of 1.226kg/m since the first metric version of CP3: Chapter V: Part 2 (1970). Therefore the recommended
3
value for air density should not be used in the UK, a value of 1.226kg/m should be used instead.

2.4.5 Section 5 Wind actions

This section gives the rules for determining the wind pressures on internal and external surfaces, the wind
forces and frictional forces. The approaches used in EN1991-1-4 to calculate wind pressures, wind forces
and friction forces are essentially very similar to those used in BS6399-2. However, there are important
differences in each of these approaches as described below:

• Wind Pressure - Expressions (5.1) and (5.2) for external and internal wind pressures do not include
the equivalent of the BS6399-2 size effect factor; this factor, called the size factor in EN1991-1-4, is
only allowed for external pressures and is included in the equations for wind force.

• Wind Force - The 0.85 factor used in BS6399-2 Equation (7) to account for the non-simultaneous
effects of wind forces on windward and leeward faces is not included in the EN1991-1-4 expressions
for wind force. In EN1991-1-4 this factor is applied directly to the pressure coefficients and varies
between 1.0 and 0.85 depending on the slenderness of the building. It is recommended in clause
5.3(5) of EN1991-1-4 that this factor only be applied to vertical walls, however, this decision is left to
national choice. The UK has decided that this factor may be applied to the horizontal force
component from both walls and roofs. This is the same as the approach used in BS6399-2.

• Friction Force - The frictional coefficients used in EN1991-1-4 are the same as those used in
BS6399-2. An inconsistent approach is used in BS6399-2 whereby the friction effects on the roof are

14