127: The urban heat island (UHI) effect - an explainer

 

The urban heat island (UHI) effect.

 

The conventional wisdom is that climate change is driven by rising carbon dioxide (CO₂) levels in the atmosphere, and only by CO₂; so the greater the (CO₂) levels the greater the temperature increase (see Fig. 87.3 in Post 87). You may have noticed one direct consequence of this orthodoxy in the way the media these days reports on environmental disasters or extreme weather (floods, droughts, storms, hurricanes, heatwaves, forest fires etc.): they always refer to climate change.

Implicit to this climate change reference is the assumption that all climate change is due to CO₂ even though CO₂ is rarely explicitly mentioned and the causation is rarely demonstrated. Consequently, the solution to all extreme weather events appears to be simple and obvious: cut CO₂ levels in the atmosphere (i.e. Net Zero) and everything will be fine. Except it won't. This is because much of what is happening to local climates has little or nothing to do with CO₂, but it does have a lot to do with other human activities, not least urbanization and industrialization. Central to both of these is the urban heat island (UHI) effect.

The problem when discussing the impact of the UHI effect on climate, and in particular the temperature record, is that it is controversial. This is partly because much of climate science appears to be driven by an anti-fossil fuel dogma that therefore sees any talk of UHIs as at best a distraction from the supposed only true problem, CO₂, and at worst a campaign of disinformation designed to undermine all of climate science and its campaign against CO₂. But it is also partly because UHIs come in many flavours. 

There are those UHIs that just trap more heat by reducing airflows and those that store more solar heat than rural areas by virtue of increased heat capacities. Both of these do not add to the total amount of energy absorbed at the Earth's surface though, so there is no net global temperature increase associated with them. But then there are those UHI processes that do absorb extra heat, either via changes to the albedo of the Earth's surface, or by the emission of large amounts of additional heat through anthropogenic energy use and generation. Both of these certainly do add to global warming but are still largely ignored by climate science. In the following sections I will discuss the relative impact of each of these four types of UHI effect in turn and show that one type in particular can be very significant.

i) Heat trapping

The aspect of the UHI effect that is referred to the most is heat trapping. This is where tall buildings in a city reduce the flow of hot air away from the centre causing the city to retain its heat longer. Perhaps the most obvious example of this is Manhattan in New York City with its dense cluster of tall skyscrapers.

The result of this UHI effect is that the local area of the city stays hotter for longer compared to if the buildings were not there. This is because there is less diffusion of heat to outlying areas, so those areas are less likely to be warmed by the city and the city is less likely to be cooled by heat transfer to the rural areas that surround it. 

However, this does not lead to more global warming because while the city will be hotter for longer than otherwise expected, the surrounding area will be cooler for longer as well because less heat from the urban areas reaches the rural areas. The key point here is that no extra heat is created at the surface of the Earth, it is just prevented from diffusing to colder regions. So the net effect on local mean temperatures is zero. As an example consider the Grand Canyon. It will trap heat in the same way that tall buildings do, but does that mean that it is warming faster than the rest of Arizona? No, and nor does it make Arizona as a whole get any warmer.

This is one reason why climate scientists discount the UHI effect, and in this case they are right, provided that the weather stations used to monitor temperature changes are evenly distributed and their temperature readings are not adjusted. Those, unfortunately, are big IFs, because any bias in station numbers between urban and rural regions compared to their relative areas will affect the the relative contribution of each to the mean global temperature, and we do know that station densities are generally higher in urban areas. So potentially there are more warm urban stations contributing to the global average than there should be and fewer cold rural ones. In an ideal world, though, this should not occur, and so neither would any contribution to global temperatures.

Net effect on global warming: zero.

ii) Increased heat capacity

Probably the second most cited variation of the UHI effect is heat retention where cities heat up and store energy from the Sun during the day and then gradually release it overnight. The net effect of this is that the maximum temperature in the city during the day should be less than expected because of the time it takes the buildings to heat up. This is because the Sun is not just heating up the top layer of the Earth's surface, as would be the case in rural areas; it is also having to heat up large concrete structures with much higher heat capacities. The higher the local heat capacity of these structures, the longer it takes to warm them and the slower, and therefore lower, their temperature rise will be. This in turn means that less infra-red radiation is then radiated back into outer space during the day because the region is cooler than it would be without the buildings, and so there is less heating of the lower atmosphere and less downwelling radiation.

At night, however, the heating from the Sun stops. The rural areas cool quickly but the urban areas don't because the urban areas have the much higher heat capacity: there is more heat stored that needs to be lost before a new thermal equilibrium without the Sun can be established. So the buildings are now warmer than their rural surroundings but are slowly cooling, acting like large radiators or storage heaters. This means that the city stays warmer for longer, and temperatures within the city are higher at night than they would otherwise be. 

The net effect of this is that temperatures during the day will be lower, but those at night-time will be higher. Overall, though, the effect on the average temperature will be zero as the two changes in temperature cancel due to the fact that the changes in heat absorption will also cancel.

Net effect on global warming: zero.

iii) Increased heat absorption

One consequence of urban development is that it changes the reflectivity of the Earth's surface for incident visible, ultraviolet and near infra-red radiation. This reflectivity is known as the albedo and it is loosely related to the colour of a surface: darker colours tend to absorb more radiation while lighter ones generally reflect more. If the albedo increases, then more radiation is reflected back into space without heating the planet, so ice and snow help to cool the planet (their albedo is over 80%) while dark soil and oceans tend to warm the planet (see Table 14.1 in Post 14 for a list of typical albedos). It therefore follows that if the colour of a surface changes, then so will its albedo, and this can then change the amount of radiation absorbed at the surface. If this absorbed radiation increases, then the Earth will get warmer and the UHI effect is one way this can happen.

In Post 14 I explained that of the average incoming solar radiation of 341 W/m² that the Earth receives, only 161 W/m² is absorbed at its surface, and that greenhouse gases then amplify this with 333 W/m² of additional downwelling radiation. This total absorbed heat of 494 W/m² then dictates the mean surface temperature via the Stefan-Boltzmann law (see Post 12). It therefore follows that if any change occurs at the Earth's surface that increases the 161 W/m² of absorbed radiation, then this will change the downwelling radiation by the same percentage and therefore change the mean surface temperature as well.

The process of urbanization inevitably involves changing the colour and texture of the Earth's surface. It generally means that areas of vegetation are replaced with tarmac and concrete. Buildings with dark roofs absorb more solar radiation than trees and grassland. However the situation is not straightforward because concrete can be very reflective and arable land tends to be very dark. Overall though, there is generally a small decrease in albedo with urbanization, and therefore a small increase in the amount of solar radiation that is absorbed. This will raise the surface temperature of the Earth slightly as well, but because it is small it is not likely to be significant.

One human innovation that can have a big impact on temperature is solar power. Because solar panels are designed to absorb 99% of solar radiation, they will add additional heating to any area where they are installed by reducing the albedo to less than 1%. So they may save on CO₂ emissions but they come with their own drawbacks, particularly if you live near them. And if they are added to roofs of buildings in cities and urban areas, they will substantially warm those areas.

Net effect on global warming: small increase in local temperatures.

iv) Heat production

There is one UHI effect that does significantly affect temperatures though: waste heat. This is where human energy use ends up as waste heat that heats the local environment around where the energy is being used. As I showed first in Post 14 and later in Post 29, this direct anthropogenic surface heating (DASH) can warm suburbs, cities and even whole countries by up to 1°C. But in fact even that warming is small compared to large cities like London. 

In 2013 the total energy use in Greater London from all sources was estimated at over 150,000 GWh. That is equivalent to an average power consumption of over 15 GW throughout the year. As the area of Greater London is about 1569 km², this amounts to a constant power density of 9.6 W/m². In Post 14 I explained how increasing the 161 W/m² of solar radiation absorbed by the Earth's surface by 2.25 W/m² would be sufficient to increase the mean surface temperature by 1°C. But I also explained that any other source of heat that was absorbed or produced at the surface would have the same effect. So 2.25 W/m² of waste heat generated at the surface would also lead to 1°C of warming.

In London the waste heat will amount to 9.6 W/m², more or less the same as the total power usage. This is because, according to the second law of thermodynamics, all energy is destined to end up as heat or entropy eventually. So waste heat is probably responsible for over 4°C of warming in London - not a great shock to people who live there. That is the urban heat island effect (UHI).

Of course not everyone sees it this way. In climate science this warming is dismissed as trivial because it only amounts to 0.028 W/m² of power use when averaged across the entire surface of the Earth, and so it only raises global mean temperatures by about 0.01°C. While this is technically correct, it neglects the uneven distribution of both these heat sources and the weather stations that determine the global temperature. Most weather stations are on land, almost 90% are in the Northern Hemisphere, and most of these are in the USA, Europe and China. So a high proportion are going to be distorted by the UHI effect from waste heat. That is what makes it important.

Net effect on global warming: large increase in large cities and much of Europe and the USA.

Summary

What I have shown here is that most types of urban heat island (UHI) have little or no effect on global warming with one exception: waste heat. This can add several degrees to the local temperatures.

However, even this is not the full story because the existence UHIs of themselves is not the only issue. Just because a small area of the Earth's surface retains or produces more heat than another does not mean that overall temperatures will rise and add to global warming. It is the change in heat retention and emission over time that is important, not the magnitude or difference from the rest of the environment. A UHI has no impact on global warming if its energy usage is not changing over time. Unfortunately in most cases the energy usage has changed, and by a large amount.

In the next six posts I will highlight six extreme examples of UHIs in the Southern Hemisphere. These are all examples of UHIs in large cities where the UHI temperature has increased much faster than that seen in the country or region as a whole, probably due to significant growth in the size, population and energy use in those cities.

29. Lateral thought #1 - suburban heating

Question

How much does the average home heat up its environment?

This is really a question that ties in with what I wrote on Post 14. Surface Heating, but I think it illustrates the point at a level that most people can relate to.
 

Answer

Well, we know from Trenberth et al. that the average power of solar radiation incident on the Earth's surface is approximately 161 W/m² (see Fig. 14.1). We also know that this leads to a mean surface temperature for the Earth of about 288 K. We also know from Post 12 (black body radiation and Planck's law) that the emitted surface radiation density scales as T ⁴, where T is the absolute temperature of the surface measured in kelvins, and the emitted radiation must balance the incoming radiation. In other words, both incoming and outgoing surface radiation densities will be proportional to T ⁴. For the outgoing radiation the constant of proportionality will be the Stefan-Boltzmann constant, while for the incoming radiation it will be the Stefan-Boltzmann constant divided by the feedback amplification factor. It therefore follows that if the mean surface temperature were to increase by 1 K to 289 K, then the quantity T ⁴ must increase by 1.40 %. And there are two main ways that this could be achieved. 

The first is to increase the feedback or radiative forcing through an increase in the strength of the Greenhouse Effect. This is what most climate scientists concentrate on, and what they believe is responsible for any temperature changes. But the second possibility is to increase the radiation power absorbed by the surface before the feedback amplification occurs. This could happen if the strength of the Sun's output changes, but more realistically it will happen whenever extra heat is liberated at the surface of the Earth. The amount of heat required to do this will be 1.40 % of 161 W/m², or 2.25 W/m². So an increase of 2.25 W/m² in the incident surface energy density will result in a 1 °C temperature rise (see Post 13 - Case 2).

As I pointed out in Post 14, a major source for such additional heat liberation at the Earth's surface is energy generation and consumption by humans, often for industrial needs. This leads to direct anthropogenic surface heating (DASH) that can raise the temperature of whole countries by as much as 1 °C. But it is not just industry that can significantly heat the local environment.

Consider a typical home. The average household in the UK uses at least 10,000 kWh of energy per year. That equates to an average rate of usage of energy of 1.14 kW throughout the year.

The average land area of homes in the UK is at most 500 m². Most modern housing developments have more than 30 new homes per hectare (see PPG3 guidance paragraphs 57-58); older suburban developments are generally a lot less dense than this; inner city flats and terraced houses are clearly a lot more.

All of this means that the power density for heat escaping from homes will be at least 2.28 W/m² (i.e 1140 ÷ 500). In other words, the energy used by a typical household is more than sufficient to increase the local surface temperature by more than 1 °C. And remember, all this heating has got nothing to do with CO₂ emissions. Nor does this calculation include the energy consumption of commercial buildings, industry or transport.


Conclusion

The energy used by the average household in the UK each day raises the temperature of their local environment by at least 1 °C compared to pre-industrial levels. That will be true irrespective of the source of the energy. Renewables will not help. Nor will cutting your level of CO₂ emissions. This is all down to heat, entropy and thermodynamics.

14. Surface heating

The principal claim made by climate scientists is that global temperatures have increased by about 1 °C over the last 100 years. In the last post I outlined three ways that this might happen. The first, which was due to changes in the amount of solar radiation reaching the Earth, I discounted due to a lack of evidence or plausible mechanism. The last, changes to the radiative forcing term I will discuss at a later date. In this post I will consider the second possibility: changes to the amount of direct heat absorption at the surface of the Earth. There are essentially only two ways this can happen: (i) through changes to the Earth’s reflectivity or albedo; (ii) by direct heating of the surface from energy sources other than the Sun.

(i) Changing the Earth’s albedo.

As I explained in the last post, one way that the Earth's surface temperature might change is if the proportion of light from the Sun that is reflected from the surface were to change. The amount reflected is called the albedo. This effect can be seen in Fig. 14.1 below which is taken from a 2009 paper by Kevin Trenberth, John Fasullo and Jeffrey Kiehl (Bull. Amer. Meteor. Soc. 90 (3): 311–324). On the left of Fig. 14.1 where the direct radiation from the Sun (in yellow) impacts the surface, the radiation is partially reflected with 23 W/m² being reflected and 161 W/m² is absorbed. This equates to an albedo of 0.125 ( = 23/(23+161) ).

As an aside: it seems slightly suspicious that the fractions reflected at the surface (1/8) and at the top of the atmosphere (102/341 = 30%) are so close to simple fractions. Does this indicate a high degree of uncertainty in these numbers, I wonder?

Fig. 14.1: The Earth's energy balance according to Trenberth et al. (2009). 

In order for the surface temperature of the Earth to have increased by 1 °C, one way that this could have happened would be for the amount of energy absorbed at the surface to have increased over time by 2.3 W/m². If this were to be achieved through changes to the albedo, then the albedo would need to have decreased from 0.1375 to 0.125. That is a change of 0.0125. So how likely is this?

The albedo of the Earth's surface depends of the type of material of the surface, as shown in Table 14.1. It also depends on the angle of incidence of the light as light tends to reflect more off surfaces at glazing incidence. So ocean water at the equator has a lower albedo than it does near the poles. However, there is also much less surface area near the poles which consequently reduces the contribution of high angle reflectance. 

Surface	% of Earth's     Surface Area	Albedo    %	Contribution to the     Earth's Albedo
Ocean	71.00	6	0.0426
Forest	7.62	8-18	0.0091
Grassland	7.93	25	0.0198
Arable	2.37	17	0.0040
Desert sand	5.51	40	0.0220
Urban	0.21	20	0.0004
Glaciers & ice caps	2.90	80	0.0232
Shrub & tundra	2.46	15	0.0037

Table 14.1: Approximate albedo of different parts of the Earth's surface.

The most common claims made about land use and climate changes are in regard to deforestation, increasing agricultural use, and increased urbanization. First it is claimed that deforestation for farming, particularly livestock farming aids global warming. As far as changes to the albedo are concerned, the evidence in Table 14.1 seems to point the other way. Turning forests into grassland increases the albedo.

Urbanization is also generally believed to reduce albedo, partly through what is termed the urban heat island (UHI) effect. This is the theory that cities with large amounts of concrete soak up more heat, and tall buildings trap that heat. This may be true, but it may also be a small localized effect. Again the data in Table 14.1 does not support it as a major driver of global warming.

A third claim is often made about polar ice and glaciers. The claim is that, because ice and snow have high levels of albedo, any change in their total albedo would have a large impact on global temperatures. The two main negative effects cited tend to be reductions in area by melting, or black carbon soot particles that drop on the surface and reduce the albedo. The main problem here is that the changes required are huge; a 54% decrease in area, or a decrease in albedo from 0.80 to 0.37. The first obviously has not and will not happen, and the latter is very unlikely as it would require huge levels of soot deposits.

The conclusion is, therefore, that changes to the Earth's albedo are difficult to achieve, and any that might have occurred have probably produced very little real effect in terms of increasing global temperatures.



(ii) Direct anthropogenic surface heating due to human and industrial activity.

The proposition here is this. All energy generation by humans results in an output of heat or thermal energy. Not only does every industrial process produce waste heat, but all mechanical work that is done by that process eventually ends up as heat or entropy as well. These are the consequences of the Second Law of Thermodynamics, and as every physicist knows, nothing can defeat the Second Law of Thermodynamics. So as temperature is just a measure of heat and entropy, it follows that everything humans do, every industrial process they create, all the energy that goes in will, in the end, just heat up the environment.

In the last post I showed that an increase of 2.3 ± 0.5 W/m² in the amount of radiation at the surface would raise global temperatures by 1 °C. So if we can work out what the rate of energy production and consumption by humans is, then we can equate that to a global temperature rise. The starting point for this is clear: we know from IPCC reports and the protestations of climate scientists that the human race currently emits 36 gigatonnes of carbon dioxide (CO₂) into the atmosphere. That CO₂ is created primarily by three processes.

The first is the burning of pure carbon (from coal) that produces an energy output of 394 kJ/mol for the process

(14.1)

The second is burning of methane (natural gas) that produces an energy output of 882 kJ/mol for the process

(14.2)

The third is the burning of higher alkanes (from oil) that produces an energy output of about 660 kJ per mole of CO₂ for the process

(14.3)

Each of the above energy outputs is for the burning of carbon or hydrocarbons to produce one mole of CO₂. To work out how energy that amounts to in total we need to know how much of each type of fossil fuel was used.

In 2018 global coal production was 7665 million tonnes, natural gas production was 3955 billion cubic metres or 2786 million tonnes (assuming 1 cubic metre = 704.5 g), and crude oil production was 4472 million tonnes. That suggests a mean energy output of about 560 kJ/mol. As 36 gigatonnes of carbon dioxide equates to 8.18 x 10m¹⁴ moles, then the total energy consumption would have been 4.58 x 10²⁰ J for the year, or 52,268 TWh.

Fig. 14.2: Global fossil fuel consumption since 1800.

However, according to the Our World In Data website, the global energy consumption from fossil fuels in 2017 amounted to 36,704 TWh from natural gas, 53,752 TWh from crude oil and 43,397 TWh from coal (see Fig. 14.2 above). The total of these values (133,853 TWh) is 2.53 times the value based on CO₂ emissions and suggests only 39% of fossil fuel combustion results in CO₂. This higher figure equates to an average power density at the Earth's surface of 0.030 W/m² across the whole surface of the Earth. That is turn implies a global temperature increase (based on the 2.3 W/m² required for a 1 °C increase that I demonstrated in the last post) of 0.013 °C compared to pre-industrial times. This, though, still omits the impact of nuclear power and renewables.

Fig. 14.3: Global energy production by energy type (2005-2018).

According to Statistica.com renewables and nuclear energy accounted for 15.3% of global energy consumption in 2018, and fossil fuel usage in 2018 exceeded that in 2017 (see Fig. 14.3 above), so that implies a global temperature increase of at least 0.015 °C compared to pre-industrial times. This temperature increase of 0.015 °C is, however, at least 60 times less than the one the IPCC is claiming for global warming since 1850. So this suggests that any resulting surface heating is such a small effect that we can safely ignore it, right? Well, not so fast.

We know that this heat is not spread evenly, its impact is greatest in the areas where most people live and work. We know that 90% of people live in the Northern Hemisphere; we know that 99.999% of people live on land. It is also true that 90% of weather stations are in the Northern Hemisphere, and at least 99.9% of them are on land. In other words there is a high degree of correlation between where people live, where industrial energy usage is, and where the weather stations are. For example, 19.7% of the Earth's surface is land in the Northern Hemisphere. So if 90% of the energy use is found there then the mean temperature rise on land in the Northern Hemisphere will be 0.069 °C. But of course, even that fails to tell the whole story. If we look at individual countries the results become even more stark.

If we start with what has been, historically, the biggest CO₂ producer, the USA, we see that it accounts for about 20% of global energy use despite being home to only 4.3% of the world's population, and covering only 1.6% of the Earth's surface area. That suggests that the power density for surface heating in the USA should be about 0.38 W/m² (an increase by a factor of 12.6 on the global average of 0.03 W/m²). This picture is confirmed by data from the US Energy Information Administration that indicates that the total power consumption of the 48 contiguous states (excluding Hawaii and Alaska) is 100.3 x 10¹⁵ BTU (see Fig. 14.4 below) over an area of 8.08 x 10⁶ km². As 1 BTU (British thermal unit) is the equivalent of 1055 J, this gives a power density for surface heating of 0.42 W/m². Yet this increases to 0.69 W/m² in Texas and 1.11 W/m² in Pennsylvania. That means that the temperature rise in Pennsylvania due to surface heating is almost 0.5 °C. But if we look at Europe the situation is even more extreme.

 Fig. 14.4: US energy consumption since 1950 by sector (in BTU).

According to the IEA, the UK's energy usage in 2018 was 177 million tonnes of oil equivalent (Mtoe), or 2059 TWh (1 Mtoe = 11.63 MWh). As the area of the UK is only 242,495 km², that equates to a power density of 0.97 W/m² and a temperature rise of 0.42 °C. But it is safe to assume that that energy usage will not be spread evenly across the country. At least 84% of both the UK population and UK economic activity is found in England (with an area of 130,395 km²) which implies a temperature rise for England alone of 0.66 °C. Yet that is still modest compared to Belgium and the Netherlands with their much higher population densities (see Table 14.2 below) where the projected temperature rise is close to 1.0 °C. That is more than the IPCC claims for global warming from greenhouse gas emissions.

Country	Energy Usage (Mtoe)	Power Density (W/m2)	Temperature Rise (°C)
UK	177	0.97	0.42
Italy	151	0.66	0.29
France	245	0.50	0.22
Belgium	52	2.25	0.98
Netherlands	72	2.30	1.00
Germany	298	1.11	0.48
Austria	33	0.52	0.23
Switzerland	24	0.77	0.34

Table 14.2: Energy usage, surface heating and temperature rise in Europe.

What Table 14.2 illustrates is that surface heating is a significant factor in overall global warming, and it is occurring in every major EU country, including those that border the Alps. In fact the average temperature rise over all five of the main alpine countries is 0.30 °C. It is perhaps no wonder then that the alpine glaciers have been retreating for over a century, while those in Norway and New Zealand, where the population density (and also the economic activity) is much lower, have remained more stable. But what this warming is not due to is increased CO₂ levels in the atmosphere or an enhanced Greenhouse Effect. That is a completely separate issue.

The conclusion we can draw from this is that, in most developed countries, warming of up to 1.0 °C has occurred since pre-industrial times, and this warming is solely a result of industrial activity and the heat that is generated as a result of that activity. This will occur irrespective of the energy type or source used because it is the heat that is directly warming the planet, not increases in the concentration of waste gases that then add to the Greenhouse Effect. This also means that when the energy usage goes down, the temperature should go down.

This has major implications for future energy policy because it means that nuclear power and most renewables are no better than fossil fuels. It also means that the efficiency of energy generation is as important as the quantity of energy generation in determining the amount of warming.

Fig. 14.5: Efficiencies of different power sources.

As an example of the impact of energy efficiency consider the case of solar photovoltaics. The relative efficiencies of different power sources are illustrated in Fig. 14.5 above. Of these photovoltaics are among the least efficient. They are in fact only about 15% efficient, meaning that for every 100 joules of energy they harvest from the Sun, they only create 15 joules of electricity. Yet in order to do this solar cells need to be 95% efficient in terms of absorbing incoming solar radiation. In other words their albedo needs to be less than 0.05. That means that for every 100 joules of solar radiation that falls on a solar cell, 5 joules is reflected back into space, 15 joules is turned into electricity (which will then become surface heat at the point of use), and 80 joules becomes waste surface heat in the solar cell.

Now a fashionable policy proposal at the moment is to put large numbers of photovoltaics in the Sahara Desert and then pump the electricity they produce to wherever it is needed. The problem is that not only will the electricity generated heat the location of its end user, but the solar cells will heat up the desert by decreasing the local albedo from 0.40 to 0.05. That is a double whammy. It is global warming without the need for CO₂. Now you don't hear much about that from climate scientists.