Periodic Reporting for period 3 - urbisphere (urbisphere - coupling dynamic cities and climate)
Reporting period: 2023-04-01 to 2024-09-30
Climate change and urbanization are transforming life globally, with direct impacts on each other, yet they are rarely studied together across disciplines. Our aims are to forecast feedbacks between weather/climate and cities. From synergies between our disciplines (spatial planning, remote sensing, modelling and ground-based observations), we are incorporating city dynamics and human behaviour – including human vulnerability - into weather and climate forecasts/projections. We focus on effects of cities on weather and climate, variability of weather and climate within cities, within-city dynamics of peoples’ activities and how these can be up-scaled to cities globally.
Our modelling is informed by data measured with the urbisphere developed Smart Urban Observation System (SmUrObs). In yearlong deployments in the Berlin (2021-22) and Paris (2023-24) regions, SmUrbObs has measured the 3-dimensional state of the atmosphere and surface responses as it varies though the annual cycle. These data are providing both new understanding of how cities influence the atmosphere and how atmospheric events such as heat waves influence cities. This allows us to provide model evaluation datasets at unprecedented detail.
To undertake modelling, characterization of urban regions (ex: building form, materials, urban vegetation, transport routes) is critical, but a massive computational challenge on a global basis. New cloud computing techniques are being developed, and partnerships brokered to address this (ex: to ground truth estimated parameters, to identify parameter needs).
Model enhancements to improve local meteorology are being linked to new service applications (e.g. high resolution (100 m) numerical weather prediction) with dynamic feedbacks from human behaviour that modify emissions. This allows us to link how cities will impact climate change and how the impacts of climate change will influence cities, with two way feedbacks linked to the vulnerable groups impacted and adaptive capacity. The analysis couples urban form (ex: building structures) and function (ex: housing, work, recreation), and hence helps understanding of where and when vulnerable people are exposed.
The new synergies between the four PI groups allows more integrative future development strategies for cities in a changing climate to be explored. Our focus cities to date are Berlin, Paris, Bristol, London, Freiburg, Stuttgart, Heraklion, Nairobi, Beijing, Colombo and Lahore.
Our modelling is informed by data measured with the urbisphere developed Smart Urban Observation System (SmUrObs). In yearlong deployments in the Berlin (2021-22) and Paris (2023-24) regions, SmUrbObs has measured the 3-dimensional state of the atmosphere and surface responses as it varies though the annual cycle. These data are providing both new understanding of how cities influence the atmosphere and how atmospheric events such as heat waves influence cities. This allows us to provide model evaluation datasets at unprecedented detail.
To undertake modelling, characterization of urban regions (ex: building form, materials, urban vegetation, transport routes) is critical, but a massive computational challenge on a global basis. New cloud computing techniques are being developed, and partnerships brokered to address this (ex: to ground truth estimated parameters, to identify parameter needs).
Model enhancements to improve local meteorology are being linked to new service applications (e.g. high resolution (100 m) numerical weather prediction) with dynamic feedbacks from human behaviour that modify emissions. This allows us to link how cities will impact climate change and how the impacts of climate change will influence cities, with two way feedbacks linked to the vulnerable groups impacted and adaptive capacity. The analysis couples urban form (ex: building structures) and function (ex: housing, work, recreation), and hence helps understanding of where and when vulnerable people are exposed.
The new synergies between the four PI groups allows more integrative future development strategies for cities in a changing climate to be explored. Our focus cities to date are Berlin, Paris, Bristol, London, Freiburg, Stuttgart, Heraklion, Nairobi, Beijing, Colombo and Lahore.
Within urbisphere we are developing new ways to represent city dynamics in weather and climate models. Our flexible observing and data archiving system SmUrObs when deployed in Berlin (2022-23) had 24 sites (ALC, LAS, DWL, radiation/flux). Sensors then operated in Paris (2023-24), are now being installed in Bristol (2024-5). Intensive observation periods (IOPs) had mobile ALC, AWS, plus 2-3 radiosoundings at concurrent sites. Our Freiburg (2022->) 40 outdoor AWS network had 100 indoor sensors (summer 2023) measuring heat stress in a hospital and school. Optical, thermal and hyperspectral cameras deployed in Berlin, Paris and Heraklion IOPs are mapping spectral properties of city materials and facets. With satellite imagery we cover multiple scales and provide model parameter and assess weather and climate models.
Analysis of field campaign observations with high resolution modelling is helping us explore the dynamic influences of a city on its overlying boundary layer both within the city and downwind, under different conditions. The Berlin and Paris dense regional sensor networks allow us to compare inner and outer city boundary layer responses, as well as differences in air that passes parallel to the city. The campaigns allow analyses for different conditions (ex: typical spring, extreme heatwave) as vegetation (e.g. leaf-off, drought) and human activity (e.g. small areas of irrigation) across the region change roles. Land surface temperatures (LST) are supporting model evaluations across scales (NWP to building facet), benefiting from new, enhanced methods to extract and up/downscale LST. ML is used to estimate indoor and outdoor climate using projections to assess differentially impacts to people.
We undertake feedback modelling between human and atmospheric dynamics linked to (ex): energy use (ex: London), and heat stress (ex: Freiburg, Colombo, Lahore). We are enhancing understanding inter- and intra-city form and function (ex: Stuttgart, Berlin) using household surveys to examine linkages and relations between perceived heat considering urban structure types, demographics, local conditions (ex: vegetation, shaded areas) from inner to outer city.
Our weekly all research team meeting, allow all to learn about ongoing and planned work (ex: field activities) in each study city. We communicate and disseminate our work through publications, our website, GitHub, Zenodo, social media, real-time open data, press releases, policy briefings.
Analysis of field campaign observations with high resolution modelling is helping us explore the dynamic influences of a city on its overlying boundary layer both within the city and downwind, under different conditions. The Berlin and Paris dense regional sensor networks allow us to compare inner and outer city boundary layer responses, as well as differences in air that passes parallel to the city. The campaigns allow analyses for different conditions (ex: typical spring, extreme heatwave) as vegetation (e.g. leaf-off, drought) and human activity (e.g. small areas of irrigation) across the region change roles. Land surface temperatures (LST) are supporting model evaluations across scales (NWP to building facet), benefiting from new, enhanced methods to extract and up/downscale LST. ML is used to estimate indoor and outdoor climate using projections to assess differentially impacts to people.
We undertake feedback modelling between human and atmospheric dynamics linked to (ex): energy use (ex: London), and heat stress (ex: Freiburg, Colombo, Lahore). We are enhancing understanding inter- and intra-city form and function (ex: Stuttgart, Berlin) using household surveys to examine linkages and relations between perceived heat considering urban structure types, demographics, local conditions (ex: vegetation, shaded areas) from inner to outer city.
Our weekly all research team meeting, allow all to learn about ongoing and planned work (ex: field activities) in each study city. We communicate and disseminate our work through publications, our website, GitHub, Zenodo, social media, real-time open data, press releases, policy briefings.
After completing yearlong field campaigns in Berlin and Paris and long-term observations in Freiburg using SmUrObs, installation is occurring in Bristol with its complex terrain. Using the Berlin data we observe and model the dynamic linkage between surface fluxes and boundary layer growth, upwind, across and downwind of the city. The dense network of sensors, combined with high resolution (100 m) numerical weather prediction (NWP), we are able to undertake a 4D evaluation of next generation NWP, to identify development needs. The three Tier 1 cities have different characteristics, with them decreasing in spatial extent (Paris, Berlin, Bristol), have increasing topographic variability (Berlin, Paris, Bristol), have different governance histories with all having within their metropolitan region areas local governance differences (e.g. different local councils). Our network deployments have complemented both national weather services but also other research projects/groups (e.g. ERIC ICOS, WMO RDP, Meteo France, PANAME in Paris; DWD, TUB, FUTURE in Berlin).
Our modelling capacities are being expanded and different urban models linked across scales. At the end of the project, we expect that our new modelling capacities enable: a) responsive representation of various human (ex: behavioural, socio-economic, demographic) and atmospheric (ex: weather extremes, variability) dynamics, with two way feedbacks to allow identifying of spatial variability of exposure and vulnerability to climate change related hazards and key local drivers; b) provide a consistent methodology and framework to represent intra-urban effects related to climate and socio-economic modelling as well as assessment at city, regional to global scales - all with integrated uncertainties; c) be able to enhance the dynamic representation of cities within global climate models, thereby inform vulnerability and risk modelling, allowing consistent downscaling for decision making for urban risk and resilience assessments and providing more information about the dynamic nexus of exposure-vulnerability of people in the urban sphere. Our combined observations and modelling systems, will be used to create simplified urban dynamic archetypes (UDA) for these purposes.
Our modelling capacities are being expanded and different urban models linked across scales. At the end of the project, we expect that our new modelling capacities enable: a) responsive representation of various human (ex: behavioural, socio-economic, demographic) and atmospheric (ex: weather extremes, variability) dynamics, with two way feedbacks to allow identifying of spatial variability of exposure and vulnerability to climate change related hazards and key local drivers; b) provide a consistent methodology and framework to represent intra-urban effects related to climate and socio-economic modelling as well as assessment at city, regional to global scales - all with integrated uncertainties; c) be able to enhance the dynamic representation of cities within global climate models, thereby inform vulnerability and risk modelling, allowing consistent downscaling for decision making for urban risk and resilience assessments and providing more information about the dynamic nexus of exposure-vulnerability of people in the urban sphere. Our combined observations and modelling systems, will be used to create simplified urban dynamic archetypes (UDA) for these purposes.