Rapid urbanization due to population growth has reduced the greenery in high-rise, high-density cities, such as Hong Kong. The inevitable results of losing greenery are the urban heat island effect (UHE), high energy demand, environmental pollution, and greenhouse gas emissions contributing to global climate change. The transpiration and photosynthesis processes of vegetation in vertical greenery systems (VGS) help cool down the surroundings and absorb carbon dioxide. VGSs take up very little additional space, as the vegetation is grown vertically along the building envelope, which is an ideal solution for high-rise, high-density cities. Recent studies have shown that poorly maintained VGSs have a high fire risk and a tendency to cause very high heat release rates (HRR) in a case of fire. Therefore, this study focused on estimating the fire hazard of VGSs.

The fire hazard of VGSs was investigated by performing bench-scale and medium-scale experiments followed by numerical simulations. The bench-scale tests were conducted using a cone calorimeter, and the medium-scale experiments using a test rig with dimensions 100 cm × 30 cm × 3 cm (height × width × depth) specially designed to hold woodwool in the vertical plane with minimum disturbance to upward fire spread (UFS). Woodwool, which is commonly used in forest fire research as an alternative for real vegetation, models a fuel bed similar to a VGS. Twenty-eight (28) cone calorimeter tests were conducted at four cone heat flux levels, with four woodwool moisture contents (MC) (between 8% to 20%), and four bulk densities (BD) of the fuel bed (between 50 kg/m³ and 100 kg/m³). The results from the cone tests were analysed for the flammability of woodwool with MC and BD with four parameters, namely, ignitability, combustibility, sustainability, and consumability. The analysis revealed the increase in flammability with the reduction of MC and BD for all four flammability parameters. The medium-scale experiments conducted with three different MCs and BDs (within the same range as cone tests) showed that the UFS rate increases with the decrease in MC and BD. The average UFS rate was over 3.48 cm/s while the maximum temperatures along the fuel bed exceeded 700 ºC in all scenarios. The fire dynamics simulator (FDS) replicated the medium-scale experiments with a good agreement between the predicted and measured UFS rates. However, FDS model overpredicted the temperature at different heights.

A full-scale four-story building with a VGS on one side of the façade was simulated using the developed FDS model. The dimensions of the building were set as 9 m × 6 m × 12 m (width × depth × height) with a story height of 3 m. A broken window was modelled on the first floor as an opening of 1 m × 1 m, from where a fire plume emerged out due to a flash over fire scenario in the compartment. The full-scale VGS simulations show an increase in the rate of fire spread with the decrease of vegetation MC. The effect of MC on the fire spread was profound in the upward direction, where the average UFS rate was increased by 469.8%, as the vegetation MC decreased from 80% to 10%. The MC of vegetation affected both the solid phase pyrolysis and gas phase heat flow. The fire spread rates were also found to be increasing with the decrease of packing ratio (PR) of the vegetation fuel bed. The effect of PR was again quite profound in the upward direction where the average UFS rate was increased by 207.1% when the PR of the vegetation fuel bed was decreased from 0.50 to 0.10.

This study further compared the UFS hazard of VGS with two other thermally efficient building façade systems (TEBFSs), namely external thermal insulation composite system (ETICS) and double-skin façade (DSF). Three scenarios each for the three TEBFSs were modelled by varying EPS layer thickness, cavity width, and vegetation MC in ETICS façade, DSF, and VGS, respectively, to represent different simulation scenarios. Numerical simulations were conducted in FDS using the same full-scale building model and window flame ejection phenomena. Rapid UFS was observed in the VGS scenarios, recording average UFS rates of 8.97, 5.51, and 2.86 cm/s compared to the other two TEBFSs where the flame failed to reach the top of the façade within the stipulated simulation time of 300 s. The maximum temperatures reached along the façade in VGS scenarios were much higher than those in the other two TEBFSs. The fire hazard of VGSs was much higher, when not properly maintained, compared to fire scenarios of ETICS and DSF, which have been scrutinised for the fire safety in prescriptive fire codes in some countries.