Static to dynamic data mining to improve future predictions of anthropogenic heat flux
Lead Research Organisation:
University of Reading
Department Name: Meteorology
Abstract
Current modelling of anthropogenic heat fluxes are based on data sets that are representative of the past, rather than the current or future state of city energy systems. However, many of these data sets are regularly updated making it possible to update model representations of energy systems to better capture the influence of those systems on heat fluxes. As an example, satellite data can be used to highlight energy `hotspots' that can be related to anthropogenic activity at a given time and location. Incorporating data sets such as these into the modelling process can give greater accuracy in relating energy system behaviour to anthropogenic heat flux. Through incorporating dynamic data mining into the modelling process, early trends in behaviour change and system changes can be detected and investigated for better projection of current and future energy use.
Studentship Projects
| Project Reference | Relationship | Related To | Start | End | Student Name |
|---|---|---|---|---|---|
| EP/N509723/1 | 01/10/2016 | 30/09/2021 | |||
| 1687752 | Studentship | EP/N509723/1 | 01/01/2016 | 10/10/2020 | Isabella Capel-Timms |
| Description | Detailed knowledge of urban meteorology is critical for the wellbeing of city inhabitants, especially as populations continue to increase and urban effects become more prominent. Anthropogenic heat (QF) is the heat emitted by human activity through building energy use, transport and metabolism, with the magnitude from each source varying greatly across a city. Previous studies of urban meteorology demonstrate that QF can affect the near-surface air temperatures and atmospheric stability, with implications for heat stress, air pollution and weather prediction. Despite this, QF is underrepresented in weather models due to its complexity. This project has allowed for the development of a novel, agent-based QF modelling approach (Dynamic Anthropogenic activitieS impacting Heat emissions - DASH) with human behaviours governing its fundamental dynamics, integrating simple building energy and transport models. This model can be used to investigate a city's response to scenarios such as population growth or climate change. To facilitate inclusion with broader urban land surface studies and weather models, DASH is adapted to two more computationally efficient schemes in such a way that spatial, temporal and thermal response behaviours are still represented. DASH is also coupled to an urban land surface model, SUEWS, to allow for feedback with the outdoor environment, e.g. temperature response. DASH is developed for Greater London for October 2014 - September 2015 at a local resolution. It is evaluated against both an existing model and energy statistics for the same area and period. Expected diurnal, weekly and seasonal patterns are observed, along with variations due to building use and demographics. Areally weighted mean annual QF is 6.43 W m-2 in the evaluated instance, and 8.47 W m-2 when coupled to SUEWS (under different forcing meteorology). The two more efficient schemes, with parameters derived from DASH, show means of 6.68 and 7.44 W m-2, but broader distributions overall. |
| Exploitation Route | DASH could be used for multiple purposes. Output data from DASH and the integrated building energy model (STEBBS) include the energy consumption of different building types and end uses, which could find use in the energy sector for the investigation of city-scale energy consumption as a response to activity and movement patterns. Meteorologists may explore feedbacks in more detail, such as the differences seen in surface energy balance fluxes and outdoor air temperature between coupled and uncoupled runs under the same initial and forcing conditions. The DASH-SUEWS coupling could be used for heat stress studies to explore the effects of temperature increase due to QF on human health, which could find application in policy. Such studies do exist, but DASH would be able to provide simulation at fine spatial and temporal scales across a city to better highlight local inequalities and hotspots. The flexibility of agent-based modelling allows for many opportunities to study different QF scenarios. These may include changes in population, building stock, the transport network and technology. |
| Sectors | Environment |
| URL | https://gmd.copernicus.org/articles/13/4891/2020/ |
| Description | Met Office |
| Organisation | Meteorological Office UK |
| Country | United Kingdom |
| Sector | Academic/University |
| PI Contribution | Model evaluation Model development Observations |
| Collaborator Contribution | Model development Model output data |
| Impact | ASSURE - preparation for model evaluation AerFO |
| Start Year | 2007 |