Methodological advancements to improve boundary conditions for ENVI-met microclimate simulations based on incomplete measurement data

Precise boundary conditions are crucial for urban microclimate simulations because, unlike macroscale models, their smaller spatial extent makes the model’s prognostic variables more dependent on these conditions. Incoming shortwave and longwave radiation are the most critical parameters for accurate results, serving as the primary drivers interacting with various urban structures and materials within the model domain. Thus, accurately obtaining these parameters through measurements and using them as boundary conditions is essential. They are often neither directly measured nor estimated through proxies, such as cloud cover, or are only partially available (e.g., only global radiation), leading to unnecessary inaccuracies when comparing measurements with simulations.

To address this challenge, we developed a method employing missing data algorithms to deduce cloud cover at three different height levels, from which site-specific radiation data can be derived. This method enhances input accuracy by decomposing global radiation into direct and diffuse components, facilitating the construction of a robust forcing file. When longwave radiation data is unavailable, information about air temperature, humidity and cloud cover can be used to estimate incoming longwave radiation data. Consequently, our approach significantly enhances the simulation of Mean Radiant Temperature (MRT), air temperature, and humidity, aligning closely with in-situ measurements.

In this contribution, we present the methodology and demonstrate its successful application in a model area in Hong Kong. The results prove the effectiveness in generating realistic microclimatic predictions that align with empirical observations, even with incomplete datasets and in complex environments. This approach offers a valuable tool for studies facing challenges with accurate measurements, yielding more precise simulation results, especially in complex urban environments where precise data acquisition is difficult.