Gee where have we seen something like this before? Dads/Moms and Grandparents: if you’d like your children/grandchildren to be able to do something on UHI for the spring science fair, here’s an easy to do idea. – Anthony
From Science @ NASA.gov – Satellites Pinpoint Drivers of Urban Heat Islands in the Northeast
Cities such as New York, Philadelphia, and Boston are prominent centers of political power. Less known: Their size, background ecology, and development patterns also combine to make them unusually warm, according to NASA scientists who presented new research recently at an American Geophysical Union (AGU) meeting in San Francisco, Calif.
Summer land surface temperature of cities in the Northeast were an average of 7 °C to 9 °C (13°F to 16 °F) warmer than surrounding rural areas over a three year period, the new research shows. The complex phenomenon that drives up temperatures is called the urban heat island effect.
Heat islands are not a newly-discovered phenomenon. Indeed, using simple mercury thermometers, weather watchers have noticed for some two centuries that cities tend to be warmer than surrounding rural areas.
Likewise, researchers have long noticed that the magnitude of heat islands can vary significantly between cities. However, accurate comparisons have long eluded scientists because ground-based air temperature sensors tend to be unevenly distributed and prone to local bias. The lack of quantifiable definitions for urban versus non-urban areas has also hindered comparisons.
Satellite technology, which offers a more uniform view of heat islands, is in the process of changing this. The group of researchers from NASA’s Goddard Space Flight Center in Greenbelt, Md., presented results based on a new method for comparing heat islands at the AGU meeting.
| Visible Light | Surface Heat | Developed Land | Vegetation Cover |
|---|---|---|---|
| › Larger image | › Larger image | › Larger image | › Larger image |
| Visible Light | Surface Heat | Developed Land | Vegetation Cover |
|---|---|---|---|
| › Larger image | › Larger image | › Larger image | › Larger image |
Satellite-produced maps of Providence and Buffalo highlight the role that differences in development patterns and vegetation cover can have on the magnitude of a city’s urban heat island. Though the two cities have the same approximate size, Providence has a significantly stronger heat island. Credit: NASA/Earth Observatory
“This, at least to our knowledge, is the first time that anybody has systematically compared the heat islands of a large number of cities at continental and global scales,” said Ping Zhang, a scientist at Goddard and the lead author of the research.
Land surface temperatures in cities, particularly densely-developed cities, tend to be elevated in comparison to surrounding areas — a phenomenon called an urban heat island. Credit: NASA
Heat islands can be deadly. This graph shows how the number of deaths spiked in Paris during a sweltering heat wave in 2003. Credit: University of Hawaii at Manoa/Benedicte Dousset
Air conditioning systems release waste heat into the atmosphere such that their widespread use can inadvertently elevate city air temperatures. This graph shows the result of a model that calculated the likely magnitude of the effect during the 2003 heat wave in Paris. Credit: Météo France/Cécile de Munck
Surface temperatures vary more than air temperatures during the day, but they both are fairly similar at night. Credit: EPA
› Larger image Development produces heat islands by replacing vegetation, particularly forests, with pavement and other urban infrastructure. This limits plant transpiration, an evaporative process that helps cool plant leaves and also cools air temperatures, explained Robert Wolfe of Goddard, one of the scientists who developed the method.
Dark city infrastructure, such as black roofs, also makes urban areas more apt to absorb and retain heat. Heat generated by motor vehicles, factories, and homes also contributes to the development of urban heat islands.
A New View
The new method for comparing cities, which the team of scientists has honed for about two years, involves the use of maps of impervious surface area produced by a United States Geological Survey-operated Landsat satellite, and land surface temperature data from the Moderate-resolution Imaging Spectroradiometer (MODIS), an instrument aboard NASA’s Aqua and Terra satellites.
Impervious surfaces are surfaces that don’t absorb water easily, such as roads, roofs, parking lots, and sidewalks. Land surface temperatures tend to be higher and more variable than air temperatures, but the two generally vary in sync with each other.
By analyzing data from thousands of settlements around the world, the Goddard team has pinpointed key characteristics of cities that drive the development of heat islands. The largest cities, their analysis shows, usually have the strongest heat islands. Cities located in forested regions, such as the northeastern United States, also have stronger heat islands than cities situated in grassy or desert environments.
Most recently, the Goddard group has shown that a city’s development patterns — whether a city is sprawling or compact — can also affect the strength of its heat island.
By comparing 42 cities in the Northeast, they found that densely-developed cities with compact urban cores are more apt to produce strong urban heat islands than more sprawling, less intensely-developed cities.
The compact city of Providence, R.I., for example, has surface temperatures that are about 12.2 °C (21.9 °F) warmer than the surrounding countryside, while similarly-sized but spread-out Buffalo, N.Y., produces a heat island of only about 7.2 °C (12.9 °F), according to satellite data. Since the background ecosystems and sizes of both cities are about the same, Zhang’s analysis suggests development patterns are the critical difference.
She found that land cover maps show that about 83 percent of Providence is very or moderately densely-developed. Buffalo, in contrast, has dense development patterns across just 46 percent of the city. Providence also has dense forested areas ringing the city, while Buffalo has a higher percentage of farmland. “This exacerbates the effect around Providence because forests tend to cool areas more than crops do,” explained Wolfe.
Cities in desert regions, such as Las Vegas, in contrast, often have weak heat islands or are actually cooler than the surrounding rural area. Providence, R.I.; Washington, D.C.; Philadelphia, Pa.; Baltimore, Md.; Boston, Ma.; and Pittsburgh, Pa.; had some of the strongest heat islands of the 42 northeastern cities analyzed.
“The urban heat island is a relative measure comparing the temperature of the urban core to the surrounding area,” said Marc Imhoff, the leader of the Goddard research group. “As a result, the condition of the rural land around the city matters a great deal.”
Heat Island Impacts
Ratcheting up temperatures can have significant — and deadly — consequences for cities. Heat islands not only cause air conditioner and electricity usage to surge, but they also increase the mortality of elderly people and those with pre-existing respiratory and cardiovascular illness.
The U.S. Environmental Protection Agency estimates that, between 1979 and 2003, heat exposure has caused more than the number of mortalities resulting from hurricanes, lightning, tornadoes, floods, and earthquakes combined.
“It is the lack of cooling at nighttime, rather than high daytime temperatures, that poses a health risk,” said Benedicte Dousset, a scientist from the University of Hawaii who also presented data about heat islands at the AGU meeting.
Dousset recently analyzed surface temperature images of Paris and showed the spatial distribution of heat-related deaths during a sweltering heat wave in 2003. Some 4,800 premature deaths occurred in Paris during the event, and excess mortality across Europe is thought to be about 70,000.
The risk of death was highest at night in areas where land surface temperatures were highest, she found. Buildings and other infrastructure absorb sensible heat during the day and reradiate it throughout the night, but the cooling effect of evaporation is absent in cities. The lack of relief, particularly among the elderly population, can be deadly, she explained.
Ramped up air conditioning usage may have even exacerbated the problem, other data presented at the meeting suggests. Cecile de Munck, of the French Centre for Meteorological Research of Meteo-France, conducted a series of modeling experiments that show excess heat expelled onto the streets because of increased air conditioner usage during heat waves can elevate outside street temperatures significantly.
“The finding raises the question: what can we do to design our cities in ways that will blunt the worst effects of heat islands?” said de Munck, who notes also that her research shows that some types of air conditioning exacerbate heat islands more than others.
Making sure cities have trees and parks interspersed throughout the compact urban cores can also help defend against heat islands. And studies shows that painting the surfaces of roads and buildings white instead of black and creating “green” roofs that include vegetation can soften urban heat islands.
“There’s no one solution, and it’s going to be different for every city,” said Dousset. “Heat islands are complex phenomena.”
Related Links:
Beating the Heat in the World’s Big Cities
http://earthobservatory.nasa.gov/Features/GreenRoof/
EPA Heat Island Resources
Ecosystem, Vegetation Affect Intensity of Urban Heat Island Effect
www.nasa.gov/mission_pages/terra/news/heat-islands.html
Urban Heat Island: Baltimore
http://earthobservatory.nasa.gov/IOTD/view.php?id=36227
Scientific Visualization Studio: Related Materials
http://svs.gsfc.nasa.gov/vis/a010000/a010600/a010699/index.html
Briefing Materials: Slideshows
Lead Author Ping Zhang; Goddard Space Flight Center
Benedicte Dousset, University of Hawaii
Cecile de Munck; French Centre for Meteorological Research of Meteo-France
› Download pdf Adam Voiland
NASA’s Earth Science News Team
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.






Michael Cejnar says:
December 14, 2010 at 2:16 pm
“Can someone explain the oft stated claim by warmists that the satellite temperatures are similar to the surface temperatures in recent history”
NASA GISS surface tempurature is trending at .17C/decade since 1980, UAH Satelite at .14C/decade since 1978.
Satellite is lower then surface observation, surface observation are lower then ‘projections’, but alas ‘It’s worse then we thought’.
UAH
http://vortex.nsstc.uah.edu/data/msu/t2lt/uahncdc.lt
NASA Giss 14 December 201 release of latest blatherings on their ‘new more accurate method’.
http://pubs.giss.nasa.gov/docs/2010/2010_Hansen_etal.pdf
Robuk says:
December 14, 2010 at 3:01 pm
Jeff (of Colorado) says:
December 14, 2010 at 12:09 pm
True, but would be hard pressed to find a major city (except Venice) that has stayed the same size over the last 100 years. Since cities have grown over the last 100 years, then their UHI has grown and that must be subtracted from recorded temperatures to get accurate changes.
They compare urban with it`s rural neighbour to extract the UHI signal out of the urban temp when they could use rural in the first place, someone please explain why this is acceptable.
#################
well you can just look at rural sites. Here is the problem.
1. The characteristics which make a site rural are subject to interpretation.
2. The number of rural sites will be small the tighter you make your requirements.
For example: if I use the tightest requirements in terms of population,nightlights, ISA, and length of the record, then we will be reduced to something like ~1000 stations.
If they are geographical spread out, then the measure should be good.
#
#
latitude says:
December 14, 2010 at 2:35 pm
I disagree with almost everyone that has posted.
It does not matter if a city has grow or not. Does not matter how many people, cars, etc either. Airports do not matter either. None of that matters……..
All that matters is that there is no way of knowing, and no way of correcting to 1/10th and 1/100th of a degree.
######
1. You dont need to know it to that level of accuracy. Your estimate of the effect will have many digits. 2.111111111. All you really need to know is the +-;
2. we don’t have to correct for it. Basically you can estimate the size of the effect and add that to your uncertainty. This is what Jones does. Only Hansen tries to correct for it.
3. It’s not tenable to say that there is no way of knowing. At least I have not seen a formal logical proof of that. It may be hard to estimate.That estimate may have wide error bars,
Do you believe in the LIA?
Michael Cejnar says:
December 14, 2010 at 2:16 pm
Can someone explain the oft stated claim by warmists that the satellite temperatures are similar to the surface temperatures in recent history, so valiating the surface data?
#####
they are similar. they do validate the surface temps. For the most part people get confused by the “peak” claims for UHI. they hear that theUHI is 9C or 12 C. and they forget that this is agenda driven reporting that focuses on the peaks, on the heat waves.
The problem with looking at these cases is as follows.
1. Peaks are quoted to support “green” programs in cities. Like planting trees, painting roof white. Money for relief from heat waves.
2. Averages are never quoted, that would not help the alarmist agenda.
3. The existence of cool islands ( cities as heat sinks) is only briefly mentioned although their existence has been known for a long time.
4. The actual stations that get used for the data are not in the urban places that get sited. Or they are adjusted before being used, or the UHI effect is estimated as a part of the uncertainty.
When you select your stations properly and avoid dense urban enviroments, you see that the overall effect over time is mousenuts. Important mousenuts, but mousenuts.
“What matters is not how much warmer a weather station located in the City is than a weather station located outside but how much warmer that City station has become over time.”
While I appreciate the idea of that (delta) being important to the overall science of the thing, for the moment, I would settle for a readjustment of “The warmest Year Ever”.
Steven says:
“1. The peak bias you see of 7C to 9C is by no means a daily occurance.”
Uh… what do you think Steve, they save it up for when the satellite passes over ?
Just keeeding
B Buckner says:
December 14, 2010 at 1:43 pm
Lots of impervious surface area at airports where many weather stations are located, although not so many people. If the metric changes from night lights to impervious area, the UHI adjustments at airports will be large.
########
ISA, impervious surface area is calculated from nightlights and population. sorry, the regression equation specifically excluded airports:
“A model was developed to estimate the density of ISA based on the radiance calibrated nighttime
lights and the Landscan population count. The model was developed using the 30-meter USGS
Landsat derived ISA (from the NLCD-2001) as the reference data source. The 30-meter ISA was
aggregated to a one-kilometer equal area grid in an Albers projection (Figure 1). The Landscan and
radiance calibrated nighttime lights were reprojected to the same one kilometer grid (Figures 2 and 3).
Linear regression defined an equation for estimating the density of ISA. Only grid cells with
population count values of three or greater were included in the regression. This excluded airport lands
from the regression. Also excluded from the regression were outliers having extremely high
population counts (greater than 3000 persons / km2) and extremely bright lights (digital numbers
greater than 800). The regression included 470,894 grid cells.
#####
FWIW.
The various products that estimate population density have some issues with airports.
can you figure out why?
Michael Cejnar says:
Can someone explain the oft stated claim by warmists that the satellite temperatures are similar to the surface temperatures in recent history, so valiating the surface data?
**************************************
That is clearly not true. Satellite and ground station data are not absolutely known.
What is used is the anomaly [change]!
They were set equal to each other in 1978 when the satellites were launched.
Since then they have diverged significantly and over time the satellites just keep getting cooler relative to the surface stations.
http://www.woodfortrees.org/plot/uah/from:1978/to:2010/plot/uah/from:1978/to:2010/trend/plot/gistemp/from:1978/to:2010/plot/gistemp/from:1978/to:2010/trend
Sattelite
#Least squares trend line; slope = 0.0128527 per year
Surface
#Least squares trend line; slope = 0.0167461 per year
Here is the comparison from 2000 to 2010
http://www.woodfortrees.org/plot/uah/from:2000/to:2010/plot/uah/from:2000/to:2010/trend/plot/gistemp/from:2000/to:2010/plot/gistemp/from:2000/to:2010/trend
The
“NASA GISS surface tempurature is trending at .17C/decade since 1980, UAH Satelite at .14C/decade since 1978.”
Yes:
Thats .03C per decade or .3C per century. One could consider this as a limit on the
size of the UHI effect in the global mean. To be precise you have to check the sat temps over land.
The Study in Brussels estimated .7K of UHI over 180 years for a roughly similar result.
Here is a clue. if you take an all rural series for 1900-2009 and compare it with a highly urbanized set of stations, you might find a difference around ~.1C
Page 1
VOL. 16, NO. 18
15 SEPTEMBER 2003
JOURNAL OF CLIMATE
2941
Assessment of Urban Versus Rural In Situ Surface Temperatures in the Contiguous
United States: No Difference Found
THOMAS C. PETERSON
To find out how contaminated global temperature
trends were from the UHI, Peterson et al. (1999) iden-
tified each station in GHCN using both the map-based
and the satellite-based metadata. Two time series were
then created. One was the time series from the full da-
taset, the one used routinely to determine global tem-
perature trends over land areas at the National Climatic
Data Center (e.g., Lawrimore et al. 2001), and another
one produced using only data from stations that were
identified as rural by both techniques. The two time
series were very similar. The linear trend from 1880 to
1998 was 0.65C century 1 for the full dataset and the
slightly higher 0.70C century 1 for the rural-only sub-
set. The resulting conclusion was that the well-known
global temperature time series from in situ stations was
not significantly impacted by urban warming.
Why would Peterson compare the full data set which includes the cooler rural
rural data with the rural data.
The rural data showed a 0.05c higher trend, how?
http://www.ncdc.noaa.gov/oa/wmo/ccl/rural-urban.pdf.
steven mosher says:
December 14, 2010 at 1:14 pm
“The peak bias you see of 7C to 9C is by no means a daily occurance(sic).”
From Science @ur momisugly NASA.gov – Satellites Pinpoint Drivers of Urban Heat Islands in the Northeast
“Summer land surface temperature of cities in the Northeast were an average of 7 °C to 9 °C (13°F to 16 °F) warmer than surrounding rural areas over a three year period, the new research shows.”
Perhaps Mr Mosher could reconcile “peak” with “average…over a three year period”.
It only took 20 years to change the 0.1C per century assumption to 9.0C for individual cities.
The world urban population has increased from 220M in 1900 to 3,495M in 2010 or 1388%.
Now what percentage of the global temperature record is contaminated by urban heat island.
Percent Land * Percent Influenced by UHI * Average UHI Increase since 1900.
30% * 50% * 2.0C = 0.30C
Just about the amount that the satellite trends say is either UHI or unjustified adjustments made in the temperature record.
Let’s get the historical record corrected to about 0.3C or 0.4C of actual/real increase.
The crux of the surface data contamination issue lies not in the magnitude of the temperature difference between urban and rural sites at a given time , but in the magnitude of the difference at any fixed station before and after UHI or land-use changes were introduced in the area surrounding the station. Satelllite measurements cannot resolve that issue. And given the fairly unique characteristics of such changes at each station, no compensation for any specific station record can be reliably prescribed–contrary to the “homegenization” meme. It’s only in the aggregate averages that a statistical compensation for the discrepancy between urban and rural makes any sense. Such compensation turns out to be of the same order of magnitude as the putative “global warming.” That’s why no credible estimates of trends are possible where there are is no coverage by long-term (say >100yrs) uncontaminated rural station records. Such is the case form many regions of the world.
Dr A Burns says:
December 14, 2010 at 12:00 pm
The big question is exactly how much of the warming in the past century is UHI ?
If the IPCC included UHI instead of claiming it was negligible, wouldn’t the temperature graph look very different ?
A study was done in Russia on UHI in cities there. Here a video presentation of it. It’s easy to understand, In 2 parts, about 17 minutes total:
Part 1
Part 2
Mosh, the post clearly says that two “similar” cities had a huge difference – 22F and 13F. About a 10F difference.
You say you can “estimate” that and add it to your uncertainty……..
You bet you can!
noaaprogrammer says:
December 14, 2010 at 11:48 am
(In very large families with many hired farm hands, cooking from scratch meant that preparation for the next meal of the day started after cleaning up from the previous meal, so the wood stoves were going all day long.)
They grew women different back then. 🙂
Providence has a surface temperature 22 degrees F higher than the surrounding area? Doesn’t this sound preposterous? Is it one small part of Providence? It does look like the airport. I think even Dr. Phil Jones would have noticed 22 degrees Fahrenheit. I am just dumbfounded and cannot believe this is the first time the UHI has been accurately measured.
Assessment of Urban Versus Rural In Situ Surface Temperatures in the Contiguous
United States: No Difference Found
In that paper, Peterson dismisses a dozen studies showing large UHI values, because they had used raw data that had not first been HOMOGENIZED.
So this is validation of surface data by satellite?
<img
I think this shows that it is simply stupid to mix old rural records with modern urban records with satelite data with traditional data. The variables are bigger than the measured trends.
The only accurate comparisons are between weather stations that are, and have always been located, in rural areas. It is common sense.
I’m not an expert, so I was wondering what the difference is between the “land surface temperature” that they are discussing here and the typically recorded air temperature. Is “land surface temperature” the temperature measured on the actual surface? Anyone who hasever walked bare-footed knows that paved surfaces are much hotter than than natural plant-covered surfaces. And while I would not be at all surprised to find that the surfaces are 10F – 20F hotter on average in cities, I don’t think this implies the average AIR temperature is also 10F – 20F hotter.
Can anyone expound on just what is being measured by the satellites and how that correlates to the air temperature (typically measured 2 m above the ground if I am not mistaken)?
Julian in Wales says:
December 14, 2010 at 8:16 pm
I think this shows that it is simply stupid to mix old rural records with modern urban records with satelite data with traditional data. The variables are bigger than the measured trends.
The only accurate comparisons are between weather stations that are, and have always been located, in rural areas. It is common sense.
################
That’s easily done. There are many stations that started rural and remained rural.
You can easily calculate the warming from them.
You can then compare them to the satillites, and to stations that have always been urban and to those that started rural and became urban.
If you saw that rural matched the satillite, would that be useful information?
yes. its called independent confirmation.
You could even use the satillite to fill in the missing ground stations. That’s called
Odonnell 2010. The paper that was praised for rebutting Steig.
Mark Baker says:
December 14, 2010 at 6:10 pm
Providence has a surface temperature 22 degrees F higher than the surrounding area? Doesn’t this sound preposterous? Is it one small part of Providence? It does look like the airport. I think even Dr. Phil Jones would have noticed 22 degrees Fahrenheit. I am just dumbfounded and cannot believe this is the first time the UHI has been accurately measured.
########
yes mark. It’s a PEAK disturbance. During a heat wave you can always find a place that measures very hot temps. Typically in a radiative canyon. if stations were actually located at these spots. 1. it would easily spotted. 2. the QC check for outliers will catch it.
latitude says:
December 14, 2010 at 5:50 pm
Mosh, the post clearly says that two “similar” cities had a huge difference – 22F and 13F. About a 10F difference.
You say you can “estimate” that and add it to your uncertainty……..
You bet you can!
##############
its rather easy.
1. You see if those cities are even used for the global index.
2. The QC check for outliers will spot an excursion this large.
3. These spikes dont happen every day. This research is on heat waves. So, yes
every few years you get a day or two where the temp may spike. hence the power
of averaging. If a city were 22F warmer every day, it would be easy to spot. Its BECAUSE these events are rare that the scientists are focusing on ways to predict them. AGW will make heat waves more common. So now you will see models to
predict heat waves. they already have them and are testing them. Thats how
they can say that Brussels warming is 50% due to UHI.
Sky
“That’s why no credible estimates of trends are possible where there are is no coverage by long-term (say >100yrs) uncontaminated rural station records.”
largely false. But since we do have records of that length that are utterly rural, would you have a hypothesis to test?
For example, If the contaminated records show a warming of .8C what do you suppose
a randomly choosen set of long rural records will show? Make a hypothesis. Its testable.