July 5th 2022

Dave's Top Tip: Weather Files & Part L 2021

Dave's Top Tip: Weather Files & Part L 2021

Dave's "Top Tip" 12

With the release of Part L2 2021 Building Regulations, many building modellers are raising questions over which weather files to use.

My Weather Files folder currently contains 1652 weather files, and therefore, I thought a quick review and maybe a little file organisation would help point IESVE users in the right direction.

A thermal model is used to calculate the thermal performance of a building in response to its external environment (i.e. weather) and internal heat gains (e.g. occupants, lighting, and equipment). Suitable weather data must, therefore, be selected following specific requirements/guidance based on the particular building assessment.

There are many locations and types of weather data available for use in a dynamic simulation. The following information mainly references two different types of weather data; Test Reference Year (TRY) and Design Summer Year (DSY), both provided by CIBSE:

  • Design Summer Year (DSY) Used to simulate summertime overheating assessments / occupant comfort.
  • Test Reference Year (TRY) Used to calculate annual energy and carbon assessments.

Some key parameters within the weather data include:

  • dry bulb temperature (°C); wet bulb temperature (°C).
  • atmospheric pressure (hPa).
  • global solar irradiation (W·h/m2).
  • diffuse solar irradiation (W·h/m2); cloud cover (oktas).
  • wind speed (knots).
  • wind direction (degrees clockwise from North).

Building assessment protocols typically provide guidance or mandatory requirements concerning the specific type of weather data to be used.

The following is a review of calculation methodologies and discusses the appropriate weather file assignments for each.

Part L Compliance Calculations:

Previously the Building Regulations Part L 2013 compliance specification (NCM Modelling guide) stated all calculations should use the 2006 CIBSE Test Reference Years (based on weather data from 1984-2004) (Example LondonTRY05.fwt). Standard weather sets were adopted for the below sites:

  • Belfast
  • Birmingham
  • Cardiff
  • Edinburgh
  • Glasgow
  • Leeds
  • London
  • Manchester
  • Newcastle
  • Norwich
  • Nottingham
  • Plymouth
  • Southampton
  • Swindon

The most appropriate location should be chosen, normally the closest in distance, however where a particular micro-climate difference meant a different location was more appropriate, another file could be chosen.

The latest Part L regulations 2021 (released on 15th June 2022), have included an update to the required weather data (The 2016 CIBSE Test Reference Years (produced from weather data recorded from 1984-2013)). Following the same locations, the more up-to-date weather data is to be used. One thing to be aware of is the VE only allows the correct weather data in the right file reference to be selected.

The weather file reference should be defined as the Location_TRY.epw e.g “London_TRY.epw” (any inclusion of the date reference will not allow the file to be recognised. Further information found here. – The previous weather files included a date (2005) the new versions should not include this date reference – CIBSE files sometime do include so to use within the VE this has to be removed).

VE 2022 ApLocate has also been updated to include a specific Weather file for both Part L and O:



All Apache DSM’s still use the same location for the weather file assignment:

This link gives further information on the required 2016 data sets.

Contact sales@iesve.com if you wish to purchase these files. If you are a CIBSE member there is a discount available, so please remember to let your account manager know your CIBSE Membership Number.

Overheating Compliance Calculations:

There are a number of UK overheating calculation methodologies, all include different references for weather data to be considered:

Part O Compliance

The latest Building Regulations include a requirement for overheating within dwellings, this is a mandatory requirement on certain new buildings which follows the UK TM59 overheating methodology. Part O does not reference any information on weather and therefore the TM59 guidance should be followed.


TM59 gives guidance on the expected weather data which should be considered when considering occupant comfort for residential dwellings. This includes a requirement to pass the overheating criteria when the most appropriate locations DSY1 2020s high emissions, 50% percentile scenario weather data is used. These DSY weather data files include three different summer formats:

  • DSY1 – A moderately warm summer.
  • DSY2 – A summer with a short intense spell.
  • DSY3 – A summer with a longer less intense warm spell.

Each file was produced from weather recorded from 1984 - 2013, but included a different pattern to the summer period including peak temperatures and the frequency above each temperature range.

Design summer years are available for today’s standard analysis (2016 weather files) however they are also available from another batch of DSY weather files referenced as Future weather files. These future weather files include files for the same 14 location sites but cover different variables including, DSY types (1-3), emission scenarios, percentile, as well as three different time periods:

  • 2020s (2011 – 2040)
  • 2050s (2041 – 2070) and
  • 2080s (2071 – 2100)

TM59 states compliance should be achieved when considering the DSY1 2020s high emissions, 50th percentile. It also states consideration of DSY2 and 3 as well as future years (2050s and 2080s) should be considered to further test designs of particular concern. Stating a pass is not mandatory.


TM52 lists no specific weather file to be used but states an appropriate ‘design summer year’ should be used.

Miscellaneous Calculations:


The GLA guidance on overheating calculations sets the standard of weather data used for overheating as the DSY1 for the 2020s, high emissions, 50% percentile scenario.

It goes on to list a requirement to consider the possible impact of weather events on the estimate of overheating be investigated. It states mitigation measures should be present to reduce the risk of overheating during more extreme weather years (CIBSE DSY2 and DSY3).

The GLA has additional weather data specific to the location within London. This is designed to take into consideration the urban heat island effect. More information can be found within CIBSE TM49 guide.


BREEAM guidance is different in each version, previously it has been CIBSE weather data to be used when considering energy and carbon and Exeter University’s weather database ‘Prometheus’ when considering occupant comfort. Generally, each BREEAM version lists the exact weather data within the guidance and methodology sections.


Considering operational energy, especially sensitivity and scenario calculations, prediction of weather is vital, affecting all aspects of the HVAC design and building operation, picking the most relevant weather files quite often results in a large shift in whole building energy consumption.

The latest TM54 guidance does give more information on weather files. Guidance on the actual weather but also file types when considering more extreme weather, this could be for both Summer and Winter.


Whilst it’s clearly not a case of having a few weather files to run all calculations types, I think most guidance documents now list the exact weather file to be used, at least for the base case (mandatory compliance requirement) scenarios, but it’s nearly always up to the designers/client to determine what other scenarios are considered. Very rarely do I hear weather files being discussed in design meetings, maybe we should be spending more time with the clients, giving them sufficient information for them to determine how much consideration should be complete for each study. This goes for the prediction of energy and carbon, for occupant comfort but also mechanical design sizing.

Weather data can be the biggest driver for difference in building simulation results. As I have always advised - don’t trust them, test them. It’s so easy to run scenarios in the IESVE…it’s best to be safe than sorry, run scenarios, consider alternatives for both hotter and colder seasons, get meaningful data to inform decisions, and future proof your environments.

If you want more information on calculating occupant thermal comfort using the VE, please contact sales@iesve.com for more details.

You can review any of my previous Top Tips, click here

If you have any ideas or requests for Top Tips to help users get the most of VE please let Dave know: david.pierce@iesve.com