Mark Vender
CSIRO has published predictive and typical meteorological year weather files suitable for use by building energy simulation software.
For many years, building designers in Australia have faced a problem: the standard weather files for building simulation did not reflect the impacts of a changing climate, in particular summer overheating risks.
This issue was highlighted in the report Climate Change – Impact on Building Design and Energy, published last year.
“Buildings designed to be in operation 50 years from today were still actively being modelled and designed based on climate data from at least 20 years ago,” says Grace Foo, M.AIRAH, who worked on the report team. “Also, there’s a huge range of climate files that are still being used in industry, and no one really knows where they should source them.”
In response to these issues, CSIRO has created new historical and future weather files.
The dataset for the predictive files comprises 996 text files in a single zip file. Each text file contains one year of weather data in hourly intervals.
Each text file contains weather data for one of 83 Australian locations, under one of three future climate scenarios (RCP2.6, RCP4.5, or RCP8.5), for one of four future years (2030, 2050, 2070, or 2090). The RCP climate projections reflect a range of possible climate futures with varying levels of anthropogenic GHG emissions, atmospheric greenhouse gas concentrations, and the resultant radiative forcing.
The predictive weather dataset is based on a typical meteorological year of historical weather data drawn from the years 1990 to 2015. Global climate models and morphing were applied to this data to predict the future values under each climate scenario at each location.
The typical meteorological year weather files dataset comprises 83 text files in a single zip file. Each .epw file contains one year of weather data in hourly intervals for one of 83 Australian locations. The dataset is based on weather files which were developed for the Nationwide House Energy Rating Scheme (NatHERS) residential building simulation tools. The typical meteorological year weather dataset is based on historical weather data drawn from the years 1990 to 2015 using the method described by the New Zealand National Institute of Water and Atmospheric Research (NIWA).
“It’s fantastic that we now have a central repository developed by an Australian source and endorsed by the government,” says Foo. “Even better, it’s free.”
Users can download the files from the CSIRO Ag Climate Data Shop, and feed them directly into simulation software. At this stage the files are available for EnergyPlus, ESP-r and IESVE – standard energy modelling software used in Australia.
Foo says that, based on the work she did with the report team in 2019, the new data will see a stronger focus on cooling requirements and a lower requirement for heating.
“The big difference is that these updated files take data all the way up to the end of 2015,” she says. “So there’s some amount of climate change and global warming already accounted for. I wouldn’t be surprised if designers are doing direct comparisons and they find that cooling energy has increased.
“Mechanical engineers sizing plant, they might not actually see a difference in how they size, because they’re not using this data. They need to use individual weather files as opposed to a consolidated version.”
Foo explains that the next step will be for common load-calculation software, such as Camel, to incorporate the new data.
“In the meantime,” Foo says, “modellers can export the average peak cooling and heating demand and do a quick sanity check against their plants sizing.”
To download the predictive weather files, click here.
Leave a Reply