Dynamical downscaling and citizen sensing as tools in the adaptation of Swedish cities to heat

The record hot summer of 2018 and IPCC projections for Northern Europe have raised awareness of the impacts on human health and wellbeing resulting from heat stress in high latitude cities. This calls for adaptation strategies based on reliable, detailed and tailored climate data and information. Our research aims at co-creating, together with Swedish municipalities, a prototype visualization platform for enhancing the usability and relevance of climate information.

This study is part of the ongoing 4-year research project BRIGHT - “Advancing knowledge and tools for the adaptation of Swedish cities to heat”. We present the research methodology and preliminary results from (1) the dynamical downscaling of the urban climate, (2) field campaigns with low-cost thermohygrometers, and (3) the development and first results from a citizen sensing mobile application focusing on thermal comfort.

The downscaling of historical and future heat wave events from global (>25 km) to local scale (sub-kilometer) is carried out in two steps: first, the regional climate model HCLIM is used to dynamically downscale global data to 3 km grid spacing using two nested domains, and then the land surface model SURFEX is used to downscale the 2-m temperature to the grid spacing of 300 m. A novel procedure is proposed to speed up the initialization of HCLIM simulations, and we also apply the pseudo-global warming approach to get the boundary data under different warming levels for HCLIM. The optimized downscaling procedure significantly reduces the computing costs and improves the flexibility of simulations compared with the traditional dynamical downscaling method.

The last step in this process brings the climate signal down to the scale of the individual person. This is accomplished with the building-resolving radiation model Solweig, which provides mean radiant temperature fields over a detailed 3D digital surface model of the city.

Field campaigns are carried out in the cities of Linköping and Norrköping in the summers of 2022 and 2023. We monitor air temperature and humidity at 25 places in each city using HOBO and NetAtmo sensors. The locations were selected together with stakeholders, aiming at representing different local climate zones and places that are relevant for the municipalities (e.g. schools, elderly care facilities and new development areas). The analysis allowed us to quantify the intensity of the urban heat island and to identify the presence and effect of cool islands. We further employ a co-designed citizen sensing app to enable citizens, and in particular municipal employees working in elderly or child care institutions to provide their perception of heat and to comment on real-time sensor measurements.  The collected data will be analysed and compared to sensor measurements and model results to enhance the overall understanding of thermal comfort and heat in urban areas.