Doesn’t it matter what material the sunlight is falling on? Especially in terms of thermal capacitance? Can you add a temporal scale to your code? Does the Sky View Factor included the coupling between the position of the sun throughout the day and the position of the building or tree? For instance the value of shade to a particular area is much higher if it is being provided at high noon. Cool research!
Judges’ Queries and Presenter’s Replies
Presentation Discussion
-
-
The surface material does contribute to the variation of the urban heat island. I use the term albedo to describe this although albedo refers to a material’s reflectance, not its thermal properties. Thermal capacitance does have an impact. For example both water and asphalt have low albedo values, but spending time next to asphalt on a 100 degree day will be less pleasant then spending the day by a body of water. Sky view factor is temporal constant, or relatively so. Building construction and demolition can change it, as well as seasonal changes in vegetation. However, there are methods that can determine whether a location is shaded or not a particular time – this method combined with several other parameters is at the heart of solar radiation models used to estimate solar energy potential in cities like San Francisco or Boston for example. Compiling a database of these values and producing an interactive map where people can view the shading of various locations based on the time on any particular day would be possible. Thank you for your comment.
-
Thanks for the reply!
-
It was great to see a focus on the social and health impacts of Urban Heat Islands, highlighting why this should remain an issue of concern in an era of climate change.
-
Nice visualizations! It looks like some of the biggest discrepancies between modeled vs. observed are occurring in large open areas that I would guess are parks? Have you thought about accounting for the effect of evapotranspiration by including an NDVI or EVI layer in the model?
-
Further posting is closed as the event has ended.
Patricia Culligan
Faculty: Project PI
Hello: this is a very interesting research topic. As a first order predictor of local variability in urban heat island impacts, do you think sky view factor or vegetative cover is better for Iowa City, and why?
Jonathan Goergen
Thank you for the question. When I started forming my research question I was looking broadly as extreme heat as a hazard and adopted a viewpoint held by other hazard researcher that vulnerability to hazard can be described as a function of exposure, sensitivity (or social vulnerability) and coping capacity. This lead me to ponder what sort of indicators would be effective at representing the vulnerability to extreme heat and be useful for decision makers. I decided to focus on exposure and tested the efficacy of sky view factor as an indicator of exposure to extreme heat.
Upon completing my analysis, I would say the indicator dashboard approach to predicting exposure to extreme heat (which I would say is the same as the variability to the urban heat island) is ideal, as no single contributing factor to the urban heat island will effectively convey the exposure at a single location. A suite of indicators reporting a proxy for evapotranspiration, anthropogenic heat emissions, and albedo along with sky view factor would provide more information on portions of an urban area then sky view factor alone and allow for a more informed use of resources for mitigating the urban heat island. As for the specifics of your question, I think the percentage of vegetative cover or perhaps a kernel density analysis of trees would be the most effective way to indicator the level of evapotranspiration.
Catherine Gehring
Faculty: Project Co-PI
Hello Jonathan,
I am not very familiar with your topic and have a very basic question. What is the predicted relationship between sky view factor and heat hazard and how does it interact with other factors, such as moisture?
Thank you,
Catherine
Jonathan Goergen
Thank you for the question. The relationship between sky view factor and temperature is temporally dependent. Assuming an equal temperature distribution, areas with higher sky view factors will heat up faster during the day time. However, during the evening areas with high sky view factor also cool off faster. There is a lack of independence between sky view factor and moisture. Forests with their varying canopy height and lower sky view factor will heat slower and cool slower than grasslands, while the same would be true of a downtown Chicago and a mall parking lot in suburban Chicago. This is why after completing the research I presented here, I would not use sky view factor alone as an indicator of intra-urban variation to extreme heat but would use it in conjunction with other indicators that can capture information about factors such as moisture, surface albedo and waste heat generated by human activity.
Catherine Gehring
Faculty: Project Co-PI
Thank you, Jonathan.
Catherine
Liliana Lefticariu
Faculty: Project Co-PI
Hello, this is an important topic. The video was interesting but the narration was hard to hear. I am not familiar with the topic, so can you explain how the sky factor is measured and what algorithm is used to calculate it. Thank you.
Jonathan Goergen
Thank you for the question. Sky view factor is the proportion of the sky visible from a specific location. If you where standing on an infinite flat plane with no topographic variations or other man made objects, the entire sky around you would be visible and the sky view factor where you stand would be one. Conversely, if you are standing in a windowless basement, your sky view factor would be zero. Typically, sky view factor is for a single location is measured one of two ways. The first method involves taking a hemispherical or fish-eye lens photograph at a location, and simple dividing the number of pixels showing the sky by the total number of pixels. The geometric algorithm to generate the sky view factor above a single location divides the hemisphere above a location into a number of equal parts P. Starting in one direction from the location, the view angle B to the highest obstruction is calculated. The surface of obstruction, S, is calculated as (sin B^2)(360/P). Next the direction from center is rotated 360/P degrees and repeated. The summed values of all these surfaces is then subtracted from one yield the sky view factor. This can be done as a complete integral but the larger the number of directions, the more computational intensity.
My algorithm leveraged a component of a solar modeling tool that used the above approach to create a graphic modeling a hemispherical photograph above a single location. I utilized the first approach described above to calculate sky view factor for a single location, then did so iteratively across an area to calculate the sky view factor continuous across an elevation model. This process is described in the pseudo-code located on my poster.
J Yeakley
Faculty: Project Co-PI
Hi Jonathan. It seems like you have narrowed your focus well and are investigating an interesting factor as part of the urban heat island effect. I’m wondering if you might also be considering incorporating the thermal properties of the surfaces exposed to varying amounts of sky-view? That factor would also seem very important to being able to predict the effect of increased exposure on the surface heat balance. Thanks, Alan
Jonathan Goergen
This a good question. While I look forward to how I might augment and expand my research, I have thought about how to represent the walls of buildings with respect to how they absorb and re-radiate thermal energy. As is shown in my visualization, there are tools available to extract buildings from LIDAR data, so I think it is possible to manipulate the vector building data and the elevation data to estimate sky view factor on the sides of buildings. On the topic of using the thermal properties of buildings to inform energy transfer, it may be possible at the scale I am investigating, however, it may not be possible using aerial imagery and LIDAR data alone. Building roofs are often different materials then building walls so using the assumption that the roof material is likely the same as the wall material would modeling the real world poorly. Perhaps there is a way to mine oblique angle aerial images from Google Maps to classify the material used in building walls. Crowd sourcing or extrapolating field surveys could also improve the accuracy of assigning building wall materials to specific buildings. On the second part of your question I would agree; the sky view factor of a buildings wall will influence the cooling rate while the material’s aspect, reflectance and thermal properties would influence the heating rate.