Extreme temperatures present severe health risks, and the urban heat island effect tends to exacerbate these risks within cities. Through this project, we seek to develop an integrated framework for predicting extreme temperature risks in urban areas. In the short-term, we will combine the outputs of local and global climate models with satellite and ground-based sensing and probabilistic models to perform on-line updating and assessment of the risks to urban populations from extreme temperatures. In the long-term, we intend to use this framework to optimize cooling center and temperature sensor placements for Pittsburgh and the Allegheny County area.
The basic spatio-temporal model for urban temperatures proposed in the SHADE project decomposes temperature T(x,t) into a deterministic time-varying mean T(t) , a periodic spatial field ΔT(x,t) capturing daily temperature patterns in the city, and a random field ε(x,t) capturing the temperature variance.
Temperature residuals ΔT(x,t) are modeled as a zero-mean Gaussian process in two spatial and one temporal dimensions. The figure above illustrates one potential realization of this random field for a single time slice of the model.
The above figure compares the mean predicted temperature to the actual temperature field for the New York City area, using simulated temperature sensor measurements located at the black “x” points. The model developed for this project allows for real-time predictions of urban temperatures at a resolution which more complex weather models can only achieve at much higher computational cost.
The above figure compares the 95% upper confidence bounds of the predicted temperature field to the actual temperature field for the New York City area, using simulated temperature sensor measurements located at the black “x” points. This type of information would be useful for decision-makers in deciding in what areas the maximum temperature might exceed a certain threshold, and therefore where extreme heat hazard warnings should be issued.
Source of funding
NSF PREEVENTS Track 2: Collaborative Research: “SHADE: Surface Heat Assessment for Developed Environments” (PI), with M. Berges, K. Klima (Co-PIs), and E. Bou-Zeid (PI at Princeton).
Philip and Marsha Dowd Engineering Seed Fund for Graduate Student Fellowships to PhD student Carl Malings
Malings, C., Pozzi, M., Klima, K., Bou-Zeid, E., Ramamurthy, P., Bergés, M. "Optimal Sensor Placement for Urban Heat Risk Response," 13th International Conference on Probabilistic Safety Assessment and Management (PSAM 13) 2-7 October, 2016. Seoul, Korea.
Malings, C., Pozzi, M., Klima, K., Bergés, M., Bou-Zeid, E., Ramamurthy, P. "Surface Heat Assessment for Developed Environments: Probabilistic Urban Temperature Modeling," Computers, Environment and Urban Systems 66:53-64. (Elsevier). http://dx.doi.org/10.1016/j.compenvurbsys.2017.07.006 (2017).
Malings, C., Pozzi, M., Klima, K., Bergés, M., Bou-Zeid, E., Ramamurthy, P. "Surface Heat Assessment for Developed Environments: Optimizing Urban Temperature Monitoring," Building and Environment, 141:143-154 (Elsevier) https://doi.org/10.1016/j.buildenv.2018.05.059 (2018).