Regional Climate Perspectives
  • About Me
  • Blog
  • Research Highlights
  • Data
  • Fun Analyses
Blog

New Tools in Urban Climatology

8/25/2018

0 Comments

 
            I recently attended the International Conference on Urban Climate, which, quite appropriately for an event whose major themes include megacities and hot weather, was held in New York during a typically grueling August heat wave. Among the ideas and findings were a number of demonstrations of emerging research tools. Some of them represent technological breakthroughs, others approaches applied in new and innovative ways. A selection of the most exciting are highlighted in the following paragraphs.
            As more and more money is poured into (re)designing urban areas in ways that are climate-aware, how do we ensure that this money is well-spent? For example, a common strategy is to plant more street trees, but how many and where? The usual approaches, in increasing order of accuracy and price, involve expert judgment; a few field experiments, extrapolated to the entire city; or a series of climate-model runs differing only in the surface land cover. A way to get accuracy much more easily is to employ an algorithm that can quickly run through possibilities and select the optimum. Kunihiko Fujiwara from Takenaka Corporation discussed just such an algorithm, aimed at designing an optimal Tree Arrangement Priority map for a city. This means iterating through the steps of tree arrangement, calculation of surrounding temperatures, determination of the cost-effectiveness of the tree, and finally back to slightly modifying the tree arrangement to see if the cost-effectiveness improves.
​While it might at first seem like trees should be planted where it’s hottest, or where there are the fewest trees at present, the complexity of climate feedbacks means that the solution is non-trivial to derive from, say, an urban-heat-island map like the one at right. The algorithm’s final results provide rankings of blocks that would benefit the most, for the least cost. Optimization algorithms have become a hot topic in climate science, partly related to the robust research programs that investigate where wind turbines & solar panels should be placed. As is often the case, the initial efforts are made in an area where the monetization potential is clear, but then can spread to addressing market failures, in this case mortality and economic losses from urban heat and pollution.
Picture
Climatological heat islands in Portland, Oregon, showing the high resolution with which the UHI is captured from satellite but also visually differentiating the areas which are highly vegetated from those that are barren. Source: https://www.opb.org/news/article/mapping-portlands-hottest-places/
      Tianzhen Hong from Lawrence Berkeley National Lab talked about his group's development of a new feature for the EnergyPlus software program which makes it possible to simulate energy demand of every building in a city at 10-minute intervals. To do this accurately, they must take into account its occupancy, materials, geometry, and neighbors, as well as the ambient weather conditions. The underlying platform, City Building Energy Saver, allows free analysis of neighborhoods in several US cities, both as they are and with potential modifications. This tool fills an important niche, as the interactions between adjoining buildings, neighborhoods, and even cities as a whole are drawing more attention (for example, a keynote by Marshall Shepherd discussed the nascent concept of ‘urban archipelagos’, a term implying that in some areas each island affects and is affected by the others nearby).
Picture
A 'digital synthetic city' with complex terrain and the output of a high-resolution weather model coupled to it. From Garcia-Dorado et al. (2017).
            Field campaigns are endangered. At least, that’s the sense I got from hearing several people discuss the Digital Synthetic Cities approach. A leading proponent of it is Dan Aliaga at Purdue, although it has more and more practitioners.  The essential idea is to create a digital model of a city that has the same properties as a real one – the same building sizes and materials, the same thermal properties of the streets and vegetation, the same solar-radiation input – but which only exists in digital space, making it easier to study. The verisimilitude gives it a slightly uncanny movie-like or video-game-like quality, not too different from Seahaven Island in The Truman Show. Of course, creating synthetic data or a synthetic environment is nothing new, and is happening across disciplines. This speaks to the power and universality of statistics – at their core, statistics are exactly designed to serve as a layer of abstraction, to describe things such that the actual thing is no longer needed. The novelty is in the complexity and concomitant power of these digital synthetic cities to answer questions that were previously well beyond the range of feasible computation, such as understanding the causes of small-scale precipitation patterns in a particular storm. The advances this approach will bring include a newfound ability to examine details of a certain location's climate, but also to better generalize findings as new patterns are uncovered and new processes are simulated, making it easier than ever to not only say why Place A and Place B are different, but why they are similar.
0 Comments



Leave a Reply.

    Archives

    January 2022
    November 2021
    July 2021
    October 2020
    June 2020
    May 2020
    December 2019
    August 2019
    June 2019
    January 2019
    November 2018
    October 2018
    August 2018
    June 2018
    May 2018
    April 2018
    February 2018
    January 2018
    December 2017
    November 2017
    October 2017
    September 2017
    August 2017
    July 2017
    June 2017
    May 2017
    April 2017
    March 2017
    February 2017
    January 2017
    December 2016
    November 2016
    October 2016
    September 2016
    August 2016
    July 2016
    June 2016
    May 2016
    April 2016
    March 2016
    February 2016
    January 2016
    December 2015
    November 2015
    October 2015
    September 2015
    August 2015
    July 2015
    June 2015
    May 2015
    April 2015
    March 2015
    February 2015

    Categories

    All

    RSS Feed

  • About Me
  • Blog
  • Research Highlights
  • Data
  • Fun Analyses