Climate Change and Health Seminar: Urban heat islands: Theory, measurement and mitigation

October 19, 2020

Information

Dr. Xuhui Lee, Sara Shallenberger Brown Professor of Meterorology

Yale School of the Environment

September 21, 2020

ID5779

To CiteDCA Citation Guide

00:03- Alright, I think we should start,
00:05so welcome everyone
00:07and welcome to our fourth,
00:10the first seminar of the Yale Center on Content
00:15in House in for 2020,
00:18and so today we are very please that you have dr. Xuhui Lee
00:25from the Yale School of Environment.
00:27So he's the Sara Shallenberger Brown
00:32Professor of Meteorology,
00:34he's also a director of the Yale Center
00:38for The Earth Observation,
00:41he also received the 2015 award
00:45for outstanding achievement in Balm meteorology
00:49from the American Meteorological Society.
00:52So without further ado,
00:57we will have doctors. Xuhui Lee
01:01- Thank you, Kai
01:02and also thank you Rob for having me in this event.
01:09Let me see, how do I, can you see my screen Okay?
01:15- Yes.
01:16Okay good.
01:18So I'm gonna go talk about some of all the work done
01:23on Urban Heat Island.
01:26Let me see if we can turn out the,
01:30so the title of my talk is Urban Heat Island Theory
01:36Measurement and Mitigation.
01:41So somewhere in that order,
01:42let me see if I can turn off my screen here.
01:45Okaynow, that's much better
01:48and so the work I'm presenting today
01:51is really a collection of things done by folks
01:55in my lab, current members and also past members so far
02:02of my lab, some of them are actually attending this event
02:07and I noticed that this event is being recorded,
02:10that's fine with me.
02:11There are a few slides where we don't have where we can....
02:15Where I showed you a sort of unpublished results
02:18so if you'd like to, if you want to share this recording
02:21with folks, please refrain from perhaps
02:24not sharing that part to people.
02:32So many of you are familiar
02:35with this kind of projections right?
02:36Projecting for temperature into the future
02:39to the end of the century
02:40depending on whether we take the aggressive mitigation
02:46or scenario or more of a business as new your scenario
02:49we will end up with very different temperature projection
02:52in the low emissions scenario,
02:55we expect maybe 1.5 degrees of increase, decrease dialysis
02:59increase near the end of the century
03:01but in a more sort of aggressive emission scenario RCP 8.5,
03:08the projection is that four degrees of decreases of warming
03:14towards the end of the century.
03:16So that's the kinda big picture.
03:19So what I would argue is that Heat stress
03:22is actually perhaps the most,
03:23the biggest climate threat to humans
03:27in stress associated with climate change.
03:29The reason is simple
03:31that we humans are warm blooded animals,
03:34We have a biological limit we cannot overcome,
03:38so we are warm blooded,
03:40we keep our body temperature at a constant value
03:43of the property 37 degrees Celsius
03:46and in a warm climate we need to maintain
03:49a temperature differential of at least two degrees
03:53between the thick body and the skin
03:56in order to for the metabolic heat
03:57to get discredited in the environment right?
04:01So that's a physiological limit barrier
04:03we cannot overcome if conditions
04:06in such that we cannot maintain
04:08a skin temperature lower than 35 degrees
04:11then we will suffer serious health consequences
04:16even death without of course the help of air conditioning.
04:22So that's the kind of the motivation
04:24for this kind of work off
04:26and of course we know that residents
04:29in the Urban Environment,
04:31urban residents suffer an additional Heat stress
04:35due to the Urban Heat Island.
04:36This is sort of classic depiction
04:38by Jumoke of what an urban heat Island looks like.
04:42If you have a bicycle for example
04:43your attach or sensor, something I would talk about it,
04:47you'd end up with this lecture
04:49and you move across a transect from rural to urban core.
04:56You would record temperature variations such way
05:00lower temperature in outside city,
05:03as you move to the center of city
05:06yo'll register very high temperature
05:08while relative to high temperature
05:10and this difference between urban
05:13versus rural temperature temperature
05:16is really what we call Urban Heat Island
05:17or intensity of therapy to time.
05:20So that's a well accepted sort of depiction
05:24of this phenomenon
05:27and so this is added heat
05:28that urban residents would experience,
05:31and this is a sort of spatial view
05:32for urban heat island here
05:34actually in the city of New Haven,
05:36the urban unite is very patchy.
05:38I have high spots here and there and some low spots there.
05:42So the high spots in the archaea shotguns area, right?
05:47And then that's this downtown area
05:49and then near the fringe of the city
05:52where you have a lot of trees, temperature is much lower.
05:56So that's the kind of urban heat island parent
05:59that you see in New Haven.
06:02So why Urban heat island is a concern?
06:05Well, you can just simply consider
06:07a probability distribution of temperature,
06:10this is a probability distribution temperature
06:12of maybe a rural background
06:14and Urban heat Island would shift
06:16this probability distribution just by a little bit,
06:19maybe by one degrees on average, right?
06:22But that one degree of shift in the mean
06:25would actually create a serious consequence
06:28in terms of heawave frequency
06:31and let's suppose the Heatwave threshold is here,
06:34now this is per heatwave threshold
06:36beyond which we will see problems
06:39with mobility and mortality
06:42and for Rural background, rural location,
06:46this is the area under this curve
06:49is your Heatwave frequency.
06:52Now for urban land, the simple shift in mean due to our heat
06:56on it, would change that frequency a lot,
07:00we increase that frequency a lot, right?
07:03And the other thing that you should notice
07:05of course as the urban heat Island,
07:08urban residents will actually experience
07:11a record temperatures not being seen by rural residents
07:15so again rural temperature stops here on,
07:18so this is a spread.
07:20So, but in the city,
07:22you will see temperature beyond the record, right?
07:26The record registering in the background sites.
07:29So that's also another issue
07:31that we should be concerned about Bob.
07:36So that is really the motivation
07:39for why we study the theory of urban heat island
07:44and why we want to come up with strategy
07:46to mitigate urban heat island, alright?
07:50So let me switch to give you
07:52a sort of review of theory
07:54of the urban heat island phenomenon.
07:56So this traits, they can be trace back to me many years ago
08:00to team Oaks textbook,
08:02in his textbook he listed the seven causes
08:05of urban heat Island of the seven,
08:08I highlight the four causes people consider it
08:11to be the major ones.
08:13The first one is increased absorption
08:15of short-wave radiation due to urban mophology
08:20and maybe due to the color of the landscape
08:23so they're committed...
08:25The conventional wisdom
08:26is that urban land tend to trap more solar radiation
08:29so that's a source of urban heat island.
08:32A second source of urban heat island of course
08:36is very easy to understand
08:38because there's an additional heat,
08:41anthropogenic heat from anthropogenic sources
08:44from automobile driving, driving automobiles bills,
08:47converts chemical energy in fossil fuel to mechanical energy
08:51that mechanical energy eventually dissipates
08:53as heat to the environment, right?
08:56And so another important source
08:57of anthropogenic heat is a space heating.
09:00We heat our houses or use of air conditioning
09:05and they will generate heat
09:08and dissipate heat to the environment.
09:11The third course is increased sensible heat storage
09:14on buildings and other facial structures can store energy
09:20solar energy, solar radiation energy in a daytime
09:24and that then they were released
09:26that energy at night causing nighttime urban warming,
09:30and finally not a major course is decreased evaporation
09:34You know that you'll replace natural vegetation,
09:37replacing, replace trees with artificial impervious surface
09:41you reduce evaporative cooling power right?
09:43So those are the four
09:44sort of major causes of Urban heat Island
09:49and so the, we understand
09:51those concepts in a conceptual way,
09:53in a qualitative way for a long time
09:56and so what we did was with a few years back
10:00was try to quantify those causes in a quantitative way.
10:05We believe, we know only by quantifying those causes
10:08that will then lay the foundation
10:11for sensible sort of measure of how to mitigate Binky Don.
10:18So I need to sort of take a step back
10:20and introduce this theory called
10:23The theory of intrinsic biophysical mechanism,
10:27this is theory was first developer to actually,
10:29to understand how perturbation changes surface temperature,
10:35changes near surface temperature amid arm,
10:38this theory is extended to talk,
10:40to the study of urban heat Island
10:43so some key points here.
10:45So this theory,
10:46This mechanism really is concerned with the process
10:49which how surface temperature responds
10:52to external perturbation by external perturbation,
10:56I mean a number of things.
10:57It could be addition additional aerosols to the atmosphere
11:00that will block sunlight penetration
11:04and an intercept sunlight penetration.
11:07And it could also be a change of urban, change of landscape
11:12a land use change replacing say, forest, we some open-end
11:16or natural land by urban man
11:20so those are considered to be external perturbation
11:24and so he helped Bob understand this process.
11:28There are two key components to that.
11:32Why is one called a local Longwave radiation feedback?
11:36And the other one is a change in energy redistribution
11:39but in the service in the overlaying atmosphere,
11:42I'm gonna explain those two processes in a little bit,
11:47so the way it quantified the surface temperature response
11:51is really just to do this sort of experiment
11:54or numerical experiment
11:57and then it goes quantified through measurement as well
12:01to the surface response really is the difference
12:04between temperature of old state before the perturbation
12:08and a new state after perturbation.
12:10So that's what the perturbation temperature signal
12:13is really the key here and we're trying to quantify.
12:18So let's take a look at,
12:20so let's go back to the case of deforestation study, right?
12:26The interest here is motivate your part
12:29by the new trying to send whether removing trees
12:33or adding trees or warm or cool the local temperature.
12:38So I, this is my favorite numerical example.
12:43This is a actual data collected over forest in Israel,
12:48semi arid climate conditions.
12:50This is how much solar energy reaches the forest
12:53and this is how much get reflected
12:54through its albedo reflected away from the surface,
13:00some escape of course to outer space,
13:04this is just a top of atmosphere.
13:06Now if you remove the forest
13:08and replace for us with some Shrub land,
13:12shrub land is much brighter, has higher albedo
13:16and so it's a short wave radiation
13:18well reflection will increase
13:22and so naturally you would think
13:24that the temperature would go down, right?
13:26Because now you have more
13:27or less short wave trapping solar radiation
13:32and so when the surface
13:34when we undergo what we call radiative feedback
13:38because when you have low absorption solar radiation,
13:42the surface cool and therefore they will have
13:46less Longwave radiation escaping to the from surface
13:49and eventually you will establish
13:51a new radiation liberate, right?
13:54Cause that process, the longwave adjustment,
13:57it's called Longwave feedback, that's a negative feedback
14:02and so if you allow just Longwave a radiation exchange,
14:07only allow radiation exchange to occur
14:10between the surface and atmosphere,
14:13this is you can come up with a simple prediction
14:16So the change of straight away radiation is dead ass
14:20that's your perturbation signal
14:22and the change of surface temperature Delta Ts right?
14:24This is a parameter called Local climate sensitivity,
14:28that's more or less a constant the number
14:30and so in this particular numerical example
14:33you would predict by replacing for us Shrub land
14:37and you expect a coin of dot four degrees
14:40about five degrees, right?
14:42So that's an argument some people used
14:46to promote deforestation,
14:48they're saying deforestation actually maybe a good thing
14:51cause helps cool the local climate
14:56because a lot because of albedo effect,
14:58but of course that picture is not complete
15:01because in the real world,
15:03you not only how a radiative process irradiated feedback,
15:06you also have too what I called energy redistribution
15:11occurring between the surface and the atmosphere.
15:15So there are two processes;
15:16One is evaporation.
15:18Evaporation is a process
15:19where liquid water is converted to water vapor right?
15:23So that happens near, at the surface.
15:25so evaporation that will take away energy,
15:28take away late night Tiki damage that will consume energy
15:31and then when vapor gets to the top
15:34above the atmospheric boundary layer
15:36and condenses to form cloud,
15:37that energy latent heat get released.
15:40So the process is a process of energy redistribution.
15:44It reduced screwed energy, taking away energy away
15:46from the surface, and then put the energy back
15:49into the atmosphere above the boundary layer.
15:51So that's one energy redistribution process.
15:53A second energy redistribution process is connection,
15:57is really is due, is the result of an emotion result
16:02of triplet motion in the boundary layer.
16:04That process is dissipating energy from the ground
16:11to the lower atmosphere.
16:15So you can set up this kind of thought experiment
16:18to look at how the two, the processes play out, right?
16:24In this thought experiment
16:26Or you can also do this in numerical, in the motto as well.
16:31You put a forest next to an open land
16:36and the two patches of landscape are influenced
16:40by same atmospheric conditions in terms of temperature,
16:44background temperature,
16:45in terms of incoming solar radiation, long wave radiation
16:49and so basically the value
16:51that those quantities are the same
16:53across the two patches of land
16:55at this order called a Blending height
16:58which is typically taking its first mode
17:00of great height about 50 meters
17:02to a 100 meters above the surface right?
17:04And then, so in this kind of site pair analysis
17:09all a space for a time analysis that the contrast open land
17:14the contrast in temperature which an open land
17:17and the forest land is really your,
17:19is really the deforestation signal
17:21cause that's how we approach this particular problem, right?
17:27And so I don't want to get into too much
17:29of a mathematical details except to say,
17:32this is how we frame the problem,
17:36we combined what we call
17:37the one source of a model for heat transfer,
17:43surface energy balance conservation of energy at the surface
17:47to formulate our solution for surface temperature
17:50so in this One source Model heat is dissipated
17:54from the ground to Reference height
17:58and using some kind of resistance analog right?
18:01So the heat of efficiency of heat flux
18:04is really proportional to temperature difference
18:06between difference in temperature
18:07between the surface and temperature at a lower atmosphere
18:11at a per landing height.
18:12So you combine those two sort of considerations.
18:16You'll come up with a solution for surface temperature
18:21And then you do a sort of the perturbation to decide
18:25mathematically it's just,
18:26that's equivalent to differentiating this equation
18:31and so you then get perturbation signal.
18:34That's your temp deforestation signal
18:37by replacing it four of this open land,
18:38you get a temperature change,
18:40that's the temperature change mathematically
18:42and then the temperature changes
18:43then it's partitioned into three components.
18:45The first component has to do with changing albedo.
18:50I mentioned earlier using that Israel example,
18:53the second component has to do is back.
18:54The energy redistribution efficiency has changed
18:58due to a change of reference.
19:00So forest landscape is very rough and very efficient
19:05in generating triplets,
19:06It's very efficient in dissipating energy by triplets
19:09but open land, it's very smooth so it's not as efficient.
19:13So that itself will cause change in temperature
19:17and then the third component contribution
19:20is change of energy redistribution
19:23due to evaporation change or change of evaporation
19:26and that can go either way
19:28when you compare forest to open land
19:30depending a forest cover to open land
19:33depending on which one has higher evaporation potential.
19:37So that is the approach we use to study a deforestation
19:42and it later turns out that we have two prompters here,
19:46one is this local climate sensitivity prompter
19:50which is more or less constant
19:52but this prompt F is energy redistribution factor.
19:55Some people have done quite a bit of work
19:57on this prompter and turns out this prompers
19:59more like a property of the landscape.
20:03So for example, this is a study by Bright et al
20:07looking at Energy redistribution factor
20:09for different ecosystem.
20:13This is evergreen needle-leaf forest,
20:16deciduous broad-leaf forest
20:18evergreen broad-leaf forest
20:20and this is a two types of crop lands, rain fat irrigated
20:25and this is grassland.
20:26Typically when you compare a forest
20:29versus the grass open land,
20:31you find the energy redistribution factor
20:33much high for forest
20:35especially for tropical evergreen broad-leaf forest
20:39meaning that they are a disturbance,
20:44just external sort of perturbation
20:47will not change his temperature
20:48as much same perturbation occurring over grassland
20:53because over or at this kind of landscape,
20:56the energy is can be dissipated very quickly
20:59to the atmosphere
21:00and therefore is more resistant to change in temperature,
21:05and then later on TC from my lab did this calculation
21:11mapping the energy redistribution factor across the globe
21:16given the current distribution of vegetation types
21:19of course and you find a high value in tropical places
21:24and low Value elsewhere
21:25and then Nighttime value is much lower
21:28so there's, when you look at tables
21:30night contrast Daytime energy redistribution factors
21:33is much higher than at Nighttime
21:36meaning that same amount of changes
21:39of a disturbance would cause much higher response
21:43in temperature at nighttime than in the daytime.
21:46So that kind of day and night symmetry
21:48is also very important in the consideration
21:50of how land use change affects the surface temperature.
21:55So basically then we'd say okay well,
21:57let's just extend this to urban landscape right?
22:00You've sent the urban landscape now
22:02instead of contrasting for us was open ended.
22:05We are contrasting a natural land versus urban land.
22:08That's the urban heat Island signal right?
22:11And so you go through that little model you find
22:14then now you have five contributions
22:18five factors contributing.
22:19One is changing the albedo or radiation convection effect,
22:22evaporation effect changing storage
22:25and change your anthropogenic heat.
22:27So a few years ago, my former student lays out,
22:31did this attribution analysis based on
22:36this model and then did a partitioning
22:39of urban heat island intensity
22:41to and partition the urban heat Island
22:42intensinty to different factors
22:44and this is a very complex plot that maybe I should show you
22:48I tend to just read this particular diagram.
22:50This diagram is daytime
22:52urban heat island on in situation for four cities in East,
22:57Southeast United States including where we are
23:00and so this is sort of wet climate.
23:03So and this is the modis settling observed over here.
23:08He did in intensity,
23:09this a climate model calculate intensity.
23:12This is the summation of the in individual terms,
23:16individual contributions right?
23:18So in the case of cities, this part of the world actually
23:25Albedo effect is cooling
23:27so contrary to what many people believe
23:31turns out cities in this part of the country
23:35our axe is brighter than the background,
23:39but then the rural background
23:41is mostly forests are dark
23:43so the Albedo effect is cooling
23:46but so what's surprised us actually,
23:47is this connection effect right?
23:49It turns out
23:53in this this kind of climate,
23:56this region urban land is not efficient in dissipating heat
24:00than the background forest land
24:03and so as a result of loss of convection efficiency
24:07you have an obviously a lot of warming.
24:10So it's actually this loss efficiency
24:12dominates urban heat Island intensity,
24:16is much stronger than the effect
24:19of loss of evaporative cooling, right?
24:22So that's the that kind of interpretation
24:27of the based on that model
24:30and so this kind of attribution.
24:32this kind of practitioner is obviously very important
24:34when you've tried to formulate a mitigation strategy
24:38whether you want to say for example,
24:40you want to change our Albedo or change
24:44in evaporating
24:48client trees by improving evaporation.
24:51So you can use this to help determine
24:53which one is more efficient
24:54whether Albedo of change or change of gray infrastructure
24:59or tangible green infrastructure
25:01which one gives you more cooling power.
25:05And then so that study was done prior
25:10to Google earth engine
25:10not always before Google earth engine error.
25:17So we've hand picked a 60 some cities
25:21and we manually select a satellite data
25:24and that was a lot of work right?
25:26But now we Google Earth Engine
25:28the marking of Urban heat island much easier.
25:32I just want to draw your attention
25:34to the work done by TC again,
25:36he used the Google App Engine
25:39to map out basically the urban heat island
25:42for all the cities in the world.
25:45You can go to this link and you can pick any city.
25:49I can then, there's this interface allows you,
25:53this Explorer allows you to map out local urban heat Island
25:59and also variation of time change of urban heat island
26:03or the satellite air.
26:07Now let me switch gear here
26:09and speak about mitigation right?
26:12Mitigation and we know urban heat Island
26:15is not a a good thing, especially in hot weather conditions,
26:18it exacerbate the heat stress on our urban residents
26:23so we like to perhaps modified urban landscape
26:27to comeback, to control,
26:29to reduce the intensity of Urban heat island.
26:37So this is a sort of a summary
26:40of the kind of strategies that people are considering right?
26:43One strategy is white roof,
26:45you basically convert a dark roof to replace dark roof
26:49with some kind of a white shiny bright material
26:53to increase Albedo so you then cool the urban climate.
26:59The other strategy
27:00is strategy promoted by the city of Chicago you know,
27:05putting green vegetation on rooftop
27:08like indicate this case is a City Hall
27:13and a third strategy is the one
27:14that our school used is to convert a rooftop
27:18to Solar Panel to cover the rooftop with Solar Panel.
27:22The benefit there is that instead
27:24of allowing radiation
27:27to turn into heat,
27:29you actually capture solar radiation
27:31and convert some of it into electricity
27:35and therefore avoiding heating the local environment right?
27:39So that would also bring cooling benefits.
27:41It's a fourth approach is to use
27:44Street trees
27:47to help cool
27:49whenever you can
27:50wherever you can plant trees to cooll the local climate.
27:55So the question is which one is more effective, right?
28:00And if so how do you quantify that
28:04before I do give you a solid quantification,
28:07I just want to draw your attention to this case in Chicago.
28:12It turns out changing roof top albedo
28:14is not a theoretical concept,
28:17it's actually been actively promoted in many cities,
28:20city of Chicago was one of the pioneer cities
28:23promoting this idea, promoting this approach
28:26using a brighter reflective materials
28:29to help cool the local climate
28:31to help control
28:34the local urban heat Island,
28:36this is a work done by a former student
28:41of professor Ron Smith and myself.
28:44So he quantified change in urban out Albedo
28:48in Chicago after 1995, after that notorious heat wave
28:52that kills a hundreds of people
28:54and turns out we can actually,
28:56we were able to quantify change of the citywide Albedo
29:01the city over this time period,
29:03the city Albedo has increased by a little bit by 0.02,
29:09but, so you can actually quantify,
29:12this is a homework exercise.
29:14I'll ask my students to do when they do my class
29:17and this isn't in my book, sort of homework exercise
29:21you know the question ask,
29:23the question we're asking students to do is that,
29:26when the albedo,
29:27if Albedo is increased by this much estimate
29:30how much temperature reduction you get, right?
29:33So you can basically go back to that model
29:36that I presented you earlier
29:39but now the situation is much simpler,
29:41you don't need to worry
29:42about changing energy REdistribution
29:44because we have not changed urban form.
29:46We all only did,
29:47only what we did was just to change the roof of Albedo.
29:51So you have that single prompter problem
29:54and if you put numbers together,
29:56you'll find that the 0.02 Change increase in Albedo
30:00would cause a cooling on average of 1.5 degrees Celsius.
30:05That could be quite important in the event of a heat wave.
30:11Now let me share with you the pertinent results, right?
30:15So we, that in the case of Chicago,
30:18that's, what's really a local example
30:20and then we with lays work, we use climate models
30:24and in with fall, all kinds of scenarios
30:27considerations, climate consideration,
30:29climate scenarios also mitigation scenarios
30:32using our partition efforts.
30:35So this is a...
30:36Let me help you interpret this diagram a little bit.
30:37This is the condition for Mid summer day
30:43for cities in the United States average condition
30:46of all the cities in the United States
30:48not also the 60 some cities in the United States.
30:51So this is, would be the current background temperature.
30:54You get
30:56on a hot summer,
30:57at summer noontime in rural background,
31:02okay?
31:02And this is then the urban temperatures here
31:05on the current climate condition
31:07in a future climate near the end of century,
31:10the rural background will be up here
31:12and urban temperature would be up here.
31:14So we will forever residents,
31:16we were gonna expect this much of a temperature, right?
31:20We referenced to current rural background
31:24and so by implementing core roofs
31:27we are, we stay in the model,
31:29we change all the roofs to core to highly reflective roofs.
31:34We get this much of cooling,
31:36that's substacalling substantial right?
31:38Basically you raise all the urban heat Island effect
31:43and all some greenhouse effect
31:46and then we say, okay, let's plant street trees,
31:48well, there's only a limited space
31:52for planting street trees,
31:53but we planted street trees in the model anywhere we can
31:58and also we change reflect your pavements
32:01change your pavements to reflect your material.
32:04So this is what we call additive effects,
32:07it's like the IBL from mitigation wedge, right?
32:12People talk about when we talk about dealing
32:14with greenhouse mitigation here,
32:15you can use the same idea of a wedge idea
32:18to see the attitude of strategies
32:23for mitigating urban heat Island.
32:26So in this is very aggressive scenario of course
32:30we can raise all the Urban heat island
32:35and greenhouse effect.
32:36We actually have some additional cooling
32:39of course, it's highly idealized and real world,
32:42we cannot achieve this maximum cooling
32:46but it's instructive to show that indeed
32:48a core roof Australia is much more effective
32:53than street tree or reflect your payment.
32:59So spatially, this is what this looks lik, right?
33:02If you don't do
33:06any change to the urban landscape at the end of the century
33:09you will still have a lot of urban heat Island.
33:11This is circle,
33:14warm color circles
33:15indicate Urban heat island.
33:17We have a few cities that actually have cool like Island
33:22indicated by the cold color,
33:26but they never that's on average,
33:28you've got quite strong urban heat Island
33:31but if you use EPA white roof everywhere in this cities,
33:35you actually now have a cold Island almost
33:38across the whole country.
33:42This is of course in a Daytime situation
33:44but the white roof does not work as well
33:46for nighttime obviously, right?
33:47White roof works because it reflects sunlight in the daytime
33:52but at nighttime there's no sunlight took to stick off
33:54so you don't get much of a benefit at nighttime.
33:59So that still would be still is an important
34:02hurdle to overcome how do you call a nighttime temperature?
34:08The white roof would not be an effective approach for that.
34:18So that the calculation is done really theoretical right,
34:22in the theoretical calculation
34:24and we don't really get a sense
34:27of the kind of change we are calling for,
34:29the change Urban land form is really substantial.
34:33If you really want to follow this strategy
34:36I'll be implementing white roof everywhere.
34:39So for that
34:42we decided to well the triplets,
34:43do some visualization.
34:45This visualization is based on
34:48sense fly a data source
34:51sort of drawn data collected by this company
34:55over a neighborhood in a city in,
34:59I think in Switzerland
35:03and so we then use this to it to some animation.
35:06Let me see if can turn the animation over here.
35:12It does not, let me see
35:17way by control here.
35:29(indistinct) Okay there it's go
35:33So this is the current landscape, right?
35:35We're doing a fly by as if we were a bird
35:38looking at the landscape from different angles.
35:41It's a very pleasant landscape,
35:44you know, have a dark roof
35:46green lawn and street trees
36:00and then we say, okay well,
36:01we'd like to change this landscape
36:03because we are we are very concerned
36:05about urban heat Island.
36:07So we then,
36:08we can artificially digitally alter the roof material
36:12to a white shiny high albedo material
36:16and then we'd do a fly by, right?
36:41So that, this is kind of landscape
36:43we are, we'll be looking at
36:45if we do implement that white roof strategy
36:49and of course, it's this very alien landscape,
36:52we are not very used to,
36:53a lot of people criticize us for saying that
36:55because they said, this is not a pleasant landscape
36:57to a city to be in
37:00and pass maybe you wouldn't be detrimental
37:05to pilots because they can't see the ground well
37:09and maybe they will get blinded
37:10by the Brighton yourself
37:15the roof.
37:16But anyway, so that's obviously a big change we need,
37:21we will be expecting
37:23but now let me switch gear a little bit
37:27to what we are doing now.
37:28So I won't pick a criticism
37:30of the work we have been doing is that
37:33we are using surface temperature
37:34as a measure of heat stress,
37:37temperature at the surface of landscape
37:39but people obviously, this is obviously is not accurate
37:44because to measure heat stress,
37:47you need to use air temperature
37:49and furthermore heat stress is not only caused
37:52by temperature, it's also caused by high humidity.
37:56So strictly you should,
37:58we should be using some kind of combined index,
38:01index that can combine both air temperature,
38:03not surface temperature but air temperature
38:06and also air humidity
38:09so that a perspective from the thermodynamic person,
38:12turns out the best way of measuring the combined effect
38:15is to use one called Wet-bulb temperature,
38:18in meteorology,
38:19this is how we measure Wet-bulb temperature, right?
38:22So we cover the thermometer with some kind of Wet cloth
38:25allowing the surface of the thermometer
38:27to be wet all the time
38:29and so, and allow the evaporation to occur at the surface
38:34and so the temperature you imagine
38:36that this situation is Wet-bulb temperature
38:40and so that's a thermodynamic parameter
38:42that meteorologists use a lot
38:45to characterize the thermal environment.
38:48It turns out though in a hot environment
38:52sweating is obviously is a way, it's the only way actually
38:56for us to maintain low skin temperature,
39:00a person who is sweating a lot
39:03can be considered essentially a big wet bulb
39:08cause we assume the body is exposed,
39:11no clothing and the whole body is covered with sweat
39:16so analogous to a wet bulb.
39:20So then you can use wet bulb temperature to see the effect
39:26of heat stress on human body
39:29and as I said earlier
39:32to stay alive
39:33just to survive hard environment
39:36we need to maintain a two degree difference
39:38between skin and a deep body temperature
39:42so that our body can dissipate heat
39:45to the environment right?
39:46But then it turns out if the We-bulb temperature
39:49of the environment goes beyond 35 degrees,
39:51this is no longer possible,
39:53we cannot, we wouldn't be able to be able
39:56to maintain a two degree difference.
39:58Our skin temperature would be higher than 35 degrees
40:02and if we don't have air conditioning.
40:06So without air conditioning we cannot survive
40:09when external environmental temperature
40:12or Wet-bulb temperature goes beyond 35 degrees.
40:15That's really the physiological barrier
40:19the limit that you know, determines the survivability
40:23or habitability of the law of the environment.
40:27So we are knowledge high trying to come up with a strategy
40:33of studying using a wet bulb
40:36instead of the surface temperature to quantity
40:38that's undergoing a new project,
40:40it's a collaborative project happening here at Yale,
40:45it's called Biking for Science and Health
40:48and so the idea is that we can use bicycles
40:50to help out map out temperature and humidity
40:54across urban and rural landscape
40:56and use that as a way of collecting data
40:59to validate a model calculation
41:02of course
41:04the project oe the objective of this project
41:07is much broader than only measuring temperature.
41:12So the broad objective
41:13is to integrate smart sensor technology
41:16with public bicycles
41:17or maybe private bicycles as well
41:19for urban environmental monitoring
41:22so T-Mobile for scientists
41:24including professor Dubrow as part of the team
41:29and so this is that the idea right?
41:32So we, what we want to do is to convert bicycles
41:34into measurement platform either volunteer cyclist bicycles,
41:39planning to volunteer cyclist or public bicycles.
41:43So and then, the smart sensor
41:46would sense the environmental conditions
41:48temperature humidity and in the future,
41:50we also want to measure air pollutants
41:53and so the sense of what, then you turn a cyclist smartphone
41:56into some kind of geolocation
41:58and data collection device and that data can then try
42:01and get transmitted to some kind of a server to allow
42:05and then in the case of public bicycles
42:08the data will be automatically transmitted to a data server,
42:12and then the data server
42:13would then dispatch data to different users
42:18and so that's the idea.
42:20So we are having some success
42:22in terms of designing a sensor,
42:25a smart sensor for temperature humidity.
42:27This is a patch of smart temperature humidity sensors,
42:32very small and this is a picture
42:34of all this smart sensors
42:37calibrate it against commercial sensors right?
42:41(indistinct)
42:42This is, oh sorry.
42:43Before I share with you some data,
42:45this is the kind of sensor right?
42:47It's very small
42:48or this is the interface, smartphone interface
42:52and this is to give you a scale of the sensor,
42:54a cache to the bicycle handlebar
42:58and so I'll show you that the idea we have
43:00is to recruit volunteer cyclists
43:02and eventually we can also implement sensors
43:06on public bicycles
43:07but in case of volunteer cyclists
43:09we are hoping,
43:10we are defining sort of kind of data interface.
43:12This is work by TC and Yichen interface
43:16to so that when the data is sent
43:19to some kind of data center,
43:22the cyclist would receive a link.
43:26The link then allows the cyclist to view the bicycle route
43:30as well as the conditions, temperature condition
43:34and humidity and maybe in the future
43:36also air quality parameters and along the route by spiked
43:41we are still having trouble with the color scale yet
43:43but if this is the kind of general idea, right?
43:45And so you can actually look at data, put the data
43:48this kind of spaghetti plot under different map background.
43:52This is just pure simple map background.
43:55You can put it in a,
43:57you know, satellite background map background
43:59or you can put down in street map background.
44:03So this is not place still very much a work in progress.
44:07So I was up here and see if we have questions.
44:11I like leave some time to engage.
44:14I was discussion and questions.
44:17Thank you very much.
44:20- Thank you, (indistinct) for the wonderful presentation.
44:25We do have a lot of questions from the students.
44:29But if people,
44:31if you have your own questions
44:34please type your question in the chat box while
44:39Dr. Lee was answering to the students' questions.
44:42So the first question actually
44:46don't be you showed a very very interesting
44:48with us about them, why the core roofs
44:53and I had receive a lot
44:54of question from the students asking about the comparison
44:59between a white roof versus a green roof.
45:04They were particular interesting in whether,
45:08what do you think about like the disadvantage
45:11of the white roof compared to the green roof?
45:16- So my White roof is not very pleasant, right?
45:19You don't like that in your neighborhood
45:21and if I showed you with that, a drone sort of animation
45:25the landscape's not that pleasant to look at
45:28but in terms of cooling this surface climate,
45:31white roof is much much more effective than green roof.
45:34I'll tell you why, in green roof, you have to,
45:38first of all, it's very difficult to plant trees
45:41on a roof right?
45:44So trees tend to sustain evaporation much more
45:49than grass than shrubs
45:50and so, but if you just planted shrubs
45:53and grass on rooftop, you have to constantly irrigate them
45:58in order to get cooling benefit
46:00and then your irrigation is not easy
46:02especially if you have a tall buildings
46:04and think about pumping water
46:06up to the rooftop and irrigate right?
46:08So that's itself is a very energy intensive endeavor.
46:14So absence of the radiation green roof really won't do much
46:21to the local temperature
46:23but I should have knowledge of obviously green roof
46:26is much more pleasant right?
46:28It's maybe has other benefits
46:31beyond just cooling the local landscape.
46:33So that's a debate obviously that's people should,
46:38that aspect should be considered
46:41when you look at a white roof versus a green roof.
46:44So if you look at the cooling power street vegetation
46:49is more effective
46:51than green roof.
46:52So you've put green roof here,
46:54the effect is really tiny compared to a quarrel for white.
47:01- Thanks, I think we will get more questions
47:03on these from the audience,
47:04but I will move on to the other question from the students.
47:09The other questions students are wondering is like
47:13you introduce us about the concept of urban heat Island
47:17and students are wondering like a lot of the mitigations
47:22we take for the urban area that's that has also impact
47:26for the adjacent rural areas.
47:29Like if we do all these,
47:32why move
47:34in urban area,
47:36does it also like
47:38simultaneously reduce
47:41the heat exposure in the rural area?
47:45- Yeah, that's a very good question.
47:46I think, so that really the question maybe can be brought
47:50in a little bit to say that's changing urban forms
47:55whatever way does the have effect
47:57on regional climate or even global climate?
48:00Right?
48:01The answer is probably no,
48:02because we are we are talking about change,
48:07intensive changes that's
48:08but the intensive change,
48:11is only occurs in a very tiny fraction of the landscape.
48:17Urban land is what 2% of the whole terrestrial land surface
48:21and so, and in that we have intensive modification
48:25that intensive modification will manifest itself
48:29in localized response but outside of urban area
48:33that the benefit is really really not that bad.
48:38So the answer is probably, no,
48:42unless we are dealing with like a huge metropolitan region
48:46maybe in India, where you have clusters of cities,
48:50a lot of cities cluster together
48:52maybe then there, you might have some effect
48:55on background temperature.
48:59- Thanks, I think, yeah.
49:01I think if we got a follow up customer
49:03regarding the green roofs
49:05so they were asking one of your paper,
49:09The Jaw and The Shoes article,
49:12in that paper, there's mixed implementation
49:16of the white and green roofs
49:18and the given the green roofs lead
49:20to increase the evaporation
49:22and likely increase humidity with wide roofs
49:26and green roofs have under
49:30donor's state effects
49:31due to green roofs contributing to the Web-bulb temperature
49:36- Yeah, yeah, that's an excellent point
49:39and so if you take that humidity into consideration
49:43you probably don't actually,
49:46you want to avoid a green roof
49:50because green roof
49:53on one hand you will reduce the air temperature.
49:55but on the other hand, it will increase humidity, right?
50:00So the reduction air temperature could be totally erased
50:03or the effect of temperature reduction could totally raise
50:06by enhanced humidity factors.
50:09And so, and of course in this analysis,
50:12the solid dollar analysis
50:15we have not brought in the concept of wet bulb,
50:18but if we bring wet bulb into consideration
50:20that may be an argument we should consider seriously.
50:25- Yeah, I'll also from the audience
50:27a question regarding the implementing of the
50:32core roof policy,
50:34have you considered whether you paint all the roofs white
50:38or use how they are scattered
50:41painting within the city?
50:44So do you consider the difference of the painting
50:48depend all the buildings, all you does a scattered because.
50:52- So in this calculation, we except hypothetical calculation
50:57we just combine all the routes to a high Albedo material,
51:03in actual implementation I think you cannot do that
51:06because there's no point actually doing
51:09a one size fits all situation
51:10because if you have North facing roofs right,
51:13then the deflections doesn't doesn't matter as much
51:17I saw spacing roof.
51:18So maybe you need to differentiate North facing
51:20versus South facing roofs.
51:22In the city of Chicago,
51:23they actually have grades,
51:26if you have very steep roof, they ask you,
51:28they recommend certain kind of Albedo values
51:31when you have less steep roofs,
51:33they recommend other kind Albedo
51:35so he said,
51:38it's mixed of strategy.
51:42By now all lot of cities actually
51:44aggressively promoting spokes,
51:46those kinds of reflect humid roof materials.
51:50- Thanks, I guess the audience
51:53and the students are very interested in this topic though.
51:55They have accurately both the students
51:58and audience ask a question regarding
52:01have you ever considered
52:02all these heat Island mitigation matters?
52:05They may have some side effects on the air quality
52:09so how you
52:12kissing that in your own modeling?
52:15- Yeah, there's a...
52:17So people say maybe for white roof material implementation
52:21it's best to it in clean cities
52:23where there's no, air quality is not a big concern
52:27in progic cities When you put in a white roof,
52:31you can change
52:34the way that the structure of the boundary layer
52:37essentially what happens is if you have a white roof
52:40you are not heating the low atmosphere as much.
52:44You're reflecting a lot of sunlight away
52:46without us to the upper atmosphere
52:47and to the outer space, right?
52:49So what happens then is you end up
52:51with a shallow a boundary layer
52:54but there's less mixing power, less mixing volumes,
52:57so you end up with higher air pollution concentration.
53:01So that's the, it could be a serious societal effect
53:04especially imploded seedlings.
53:05So that's another,
53:11this the harm you could say perhaps caused
53:14by air quality.
53:16That's a very good point.
53:19- Thanks, another aspect of the students are wondering
53:25is like you showed a little bit about
53:28the different
53:31like riddles from the satellite,
53:32from the modeling
53:34and the students are particularly interesting
53:36in wanting these kind of modeling.
53:39So how can you actually
53:43simulate the interactions
53:45with the global warming and also all the biophysical drivers
53:51of the urban heat Island in the continent models?
53:55- Okay, so in the climate models right,
53:58they, a lot of models don't actually have
54:00what we call a city landscape that so,
54:04but the the model we use
54:06have what we call subgrid parameterization,
54:09so within each Greek cell
54:12you have different parches for that type of land
54:15so some great cells were contained
54:20urban land tile, urban tile
54:22and some would have no, if there's no urban.
54:26So this model actually can calculate
54:29within which is great cell, temperature, humidity,
54:33and so on within for each tile.
54:38So typically when you download a data though,
54:41the data is aggregate to the Greek cell
54:43that was so you don't see subgrade kind of a pattern.
54:48You don't see a subgrade pattern
54:50but we are able to re redo the calculation
54:54and retrieve data within each Greek Model grade data
54:59for vegetations tile and offer urban tile.
55:02So that essentially set up the problem
55:03for us to have to do then compare
55:06those subgrade tile data to get the urban heat Island calc
55:09apart from the climate models.
55:11That's how a client model handles landscape heterogeneity
55:15within a model grid cell.
55:19- Thanks, I think due to the time limitation,
55:22final question
55:23is the students and audience are very interested that
55:27in like, what's your recommendations for our daily life
55:31in as an individual,
55:32is it more eco-friendly to have solar panels
55:36or have a quarter of a solar.
55:39- Solar panels are very interesting, right?
55:42You need to do a very
55:44sort of a careful calculation,
55:47to look at the benefits.
55:48So solar panel dependent if it's true false
55:50for why is that you, like I said
55:53you convert a local solar radiation to electricity
55:57and in doing so, you don't heat the environment,
56:01you don't allow radiation to heat the environment
56:04but the commercial efficiency is not very high.
56:06It's not as high as reflection by core roof.
56:11So on its own, you would say the cooling benefit
56:15of solar panel is not as high as core roof,
56:18but then you have an added benefit
56:21of electricity generated by solar energy right?
56:24So you
56:27offset the demand for fossil fuel energy.
56:30So that benefits more broad
56:32modular views is you're offsetting energy demand
56:36for fossil fuel
56:37and therefore you cool the whole club global climate.
56:41So there's that, there's a benefit to that
56:44so that you need to consider both sides
56:47local Coolig versus global cooling
56:50versus and offsetting energy
56:53and so that'd be a hard subject
56:56that need to be debated, right?
56:59But I think if you are,
57:00if you want to conserve your electricity bill,
57:03if you want to reduce your electricity bill in your house
57:06that you're,
57:07the best approach is actually having a core roof.
57:11If you have at a core roof on your rooftop,
57:13then the demand for AC will be substantially reduced.
57:18You will have a lot of electricity saving in that way.
57:23That's has to be demonstrated by a lot of people actually
57:26- One fourth session.
57:28Thank you for all the insightful discussion
57:31and also the presentation
57:33and with that, I think we thanked Dr. Lee
57:37for this wonderful presentation
57:39and I thank you all for coming for our seminar.
57:44- Bye - See you guys.