Applying urban climate model in prediction mode—evaluation of MUKLIMO_3 model performance for Austrian cities based on the summer period of 2019

Abstract

Extreme heat events are natural hazards affecting many regions of the world. This study uses an example of the six largest cities in Austria to demonstrate the potential of urban climate model simulations applied in prediction mode providing detailed information on thermal conditions. For this purpose, the urban climate model MUKLIMO_3 of the German Meteorological Service (DWD) coupled with the hydrostatic numerical weather prediction model, ALARO, is used to simulate the development of the urban heat island (UHI) in Austrian cities for the summer period of 2019 with a horizontal resolution of 100 m. In addition to the evaluation of UHI predicting skills, other relevant variables, such as humidity and wind characteristics on hourly basis, are also analysed in this paper. Model evaluation confirmed that the MUKLIMO_3 microscale model had the capacity to simulate the main thermal spatiotemporal patterns in urban areas; however, a strong dependence on the input data from the mesoscale model was found. Our results showed large benefit in prediction of maximum air temperatures in urban areas, while the relative humidity predictions of MUKLIMO_3 appear to be much less plausible and show large variety of model prediction skills. Urban climate model simulations using real atmospheric conditions can facilitate better quantification and understanding of day-to-day intra-urban variations in microclimate as well as provide a basis for evaluation of the microclimate prediction skills of mesoscale numerical models with urban extensions.

Performance analysis of the urban climate model MUKLIMO_3 for three extreme heatwave events in Bern

“…In particular, days without cold air advection or precipitation and with little wind, which generally lead to the strongest heat stress during heatwaves, can be modelled realistically with MUKLIMO_3. However, days with changes in mesoscale background conditions during the simulation period lead to larger biases between modelled and measured air temperatures, which is also detected by Hollósi et al [20]. Compared to Bokwa et al [3], this study found lower Pearson correlation values between measured and modelled air temperatures.…”

“…Changes in meteorological conditions during the model run can thus not be captured by the model. Hollósi et al [20] mention that a positive or negative cloud cover bias highly influences the diurnal ranges of air temperature and relative humidity and could show that excluding simulations of cloud cover bias of more than 50 % improve the air temperature results by about 10 %.…”

“…In order to assess the risks of heatwaves in the particularly affected urban areas, knowledge about intra-urban air temperature variability is necessary at high spatial resolutions [20]. While in-situ measurements are one way to measure the UHI effect, urban climate models (UCM) are another possible way to assess air temperature patterns in cities.…”

Impact of North Atlantic Oscillation and drought conditions on summer urban heat load ‐ a case study for Zagreb

“…The applications of the model are various: estimating the effect of urbanization (Žuvela‐Aloise et al ., 2014), effect of roof albedo modifications (Žuvela‐Aloise et al ., 2018), implementation of blue and green infrastructure (Žuvela‐Aloise et al ., 2016; Geletič et al ., 2020), and estimations of heat load in a future climate (Früh et al ., 2011; Geletič et al ., 2019a; Oswald et al ., 2020; Schau‐Noppel et al ., 2020). It was also successfully used in a prediction mode for Austrian cities (Hollósi et al ., 2021).…”

Modeling the intra-urban nocturnal summertime air temperature fields at a daily basis in a city with complex topography

“…[8,9]), numerical urban climate model simulations (e.g. [10,11]), or analyses based on self-designed urban air temperature measurement networks (e.g. [12][13][14]).…”