Ronald L. Sass

Department of Ecology & Evolutionary Biology
Rice University
Houston, TX 77005

Janet Taylor moved to Houston from the Midwest as an adult. Janet lives with her husband in a quiet residential area near the Texas Medical Center. She is a physically active woman who is trying to adjust her life style to the city’s hot and humid summer climate. It is mid August and she has just returned home after an intense game of tennis on an outdoor court near her residence. In spite of constantly drinking large quantities of water during her match, she feels overheated and ready for a cold shower. Her skin temperature is over 105°F and her face is rosy-red with heat. She considers watering her outdoor plants before showering. They too appear to be drooping with heat and thirsty for a drink. The temperature in her back yard is already over 95°F and it is not even noon! Janet rejects the impulse to protect her plants in favor of her comfortable air-conditioned sunroom, at least until it is more comfortable outside. As she cools down, she wonders if she should give up tennis until the weather gets cooler. But that may be two or three months away and tennis is a social as well as a health imperative with her. "If only the temperature were 10 degrees cooler," she wishes, "I would be a whole lot happier living here."

Janet is not alone in trying to stay comfortable in the Houston summer. All but a very few homes and places of work as well as the great majority of the cars, trucks and buses of Houston are air-conditioned. People spend as little time as possible out of doors, moving as quickly as possible from one air-conditioned oasis to another. Those whose jobs require them to be in the heat take special precautions not to be adversely affected, staying in the shade as much as possible and drinking lots of liquids. The intense summer heat also contributes to increasing the amount of smog pollution in the atmosphere. The heat is not only uncomfortable, it is unhealthy.
The tragedy of Houston’s urban heat condition is that a significant portion of it is caused by the very people who live and work in the city and suffer from it. Cities have been known for some time to be noticeably warmer than the surrounding countryside, particularly on hot, cloudless, and windless days. Temperature differences of as much as five or ten degrees Fahrenheit between a city such as Houston and nearby farmlands and prairies are not uncommon. This phenomenon is called the urban heat island (UHI) effect. A city like Houston may develop its own climate, one that is different from that of the surrounding countryside. But what causes cities to be warmer than near rural areas? What does the urban microclimate mean for those of us who live in Houston and should we be concerned about it? Would it improve our quality of life to do something about it and, indeed, is it possible to make cities cooler? These are questions that will be considered in this article.

What are the various factors that determine a particular climate?

Before attacking the subject of heat islands, a quick look at some general climate concepts will be helpful to our understanding. For example, the daily temperature in Houston, similar to that of other temperate or sub-tropical locations, varies from month to month in a significant and fairly predictable way. Everyone expects an annual cycle of seasons and plans to wear lighter clothes in the warmer summer time and heavier clothes in the colder wintertime. We are familiar with the fact that people who live further north can generally expect to be colder in the winter than those of us who live further south. We in Houston accept much warmer temperatures in the summer. These everyday experiences are consistent with the concept that the seasonal position of the sun, relative to our location is a major factor in determining our local climate. The sun supplies us with a certain amount of energy and we experience a certain heating and related temperature. The average monthly energy that we receive from the sun at a particular location on Earth changes during the year and so does the average monthly temperature at that location--the two variables are correlated in time. We can see this for ourselves in the graph in Figure 1 where both the solar energy received and average daily temperature in Houston are plotted by month. For a given month the value of the temperature and solar radiation are both averaged over all the days of the month for the period from 1961 to 1990.

One notable feature of this graph is that changes in both solar radiation and temperature are closely coupled throughout the year. Note that he solar radiation peaks about 1.5 months before the temperature. The reason for this time lag will be discussed later. Now we will consider only the general change in temperature with solar radiation. As the solar radiation increases or decreases, the temperature responds accordingly. In fact, to a very good approximation, the temperature in Houston changes by 8.6°F whenever the solar energy changes by one kilowatt-hour per square meter per day. This is true when the temperature decreases as well as increases. Amazingly, this dependence of temperature on solar input is very similar for most cities in the United States; even those located much farther north than Houston. The average monthly temperature in New York City, for example, is observed to change by a similar 8.4°F when the solar energy changes by one kilowatt-hour per meter square per day. The monthly average temperatures of Houston and New York are certainly different. But then, so are the average monthly solar radiation values.

Energy from the sun heats the Earth. But the Earth does not utilize all of the energy coming from the sun. Some energy is reflected back into space without being absorbed. All of us have experienced differences in temperature as we move about the city, particularly as we move from sunlight into cooler shade. This is because clouds are very good reflectors of the sun’s rays. Under a cloud, the surface temperature may be cooler than expected because much of the sun’s energy is reflected by the clouds and never has a chance to warm up the surface of the Earth.

Some places feel hotter in direct sunlight than others do. This is true because different surfaces interact with the sun’s rays in different ways. Some surfaces are very good reflectors of the sun’s rays and therefore do not absorb enough energy to become very hot. Other surfaces become very hot—even hotter than the ambient air temperature because they are poor reflectors of the sun’s rays but very efficient absorbers of the sun’s energy. The total fraction of solar radiation that is reflected by a particular surface is called the albedo of that surface. The higher the albedo, the more energy is reflected and the less is absorbed.

I remember a day during a visit to India in 1993. I was visiting a very famous Hindu temple in New Delhi during a very hot sunny summer day. It is customary to remove one’s shoes before entering a Hindu temple building. The open court area leading into the temple was paved with a mosaic of white, red, and dark brown tiles. Walking barefoot over the tiles was a painful reminder of scientific observation. The brown tiles were so hot that I could not stand still on them without blistering my feet. The red tiles also felt hot but not painfully uncomfortable. And the white tiles felt cool and comfortable. Most people were crossing this court area by moving from one white tile to the next, much as a child would walk to school avoiding the cracks in the sidewalk in order to "not step on a crack and break somebody’s back!" Avoidance was necessary because of the different albedos of the three colors of tiles. The white tiles were very reflective, absorbing little of the sun’s energy and were cool. The red tiles had a lower albedo and were warmer and the brown tiles were very hot because they reflected very little of the sun’s rays.

The energy transfer from the sun to the Earth is an example of radiant heat transfer. There is another radiative energy transfer that is equally important to our understanding of climate but much less intuitive, namely the transfer of radiant energy from the Earth system back into space. Both transfer processes are referred to as black body radiation. It is a property of all matter to emit radiant energy. In general, hotter the bodies emit more radiant energy at shorter wavelengths.
The concept of radiant energy can be illustrated by an experience common to all. Imagine turning on a heating coil on an electric stove. If you suspend your hand a couple of inches above the burner, you soon notice your hand feeling warmer. As time passes, your hand feels hotter and hotter. At the same time the burner may begin to show a faint red glow. If you continue this experiment, the burner will eventually glow white and your hand will become very hot. This is also an example of radiant energy transfer. The burner coil is emitting radiant energy just as the sun does, only at a longer wavelength. This energy is mainly in the infrared portion of the radiation spectrum but becomes partially visible light as the temperature of the radiating body increases. The amount of energy emitted by a body also increases as its temperature increases (actually as the fourth power of its temperature so the rate of increase is quite pronounced).

The sun is very hot, and therefore it not only emits large amounts of radiant energy, but it also emits much of this energy in the ultra-violet and the visible part of the energy spectrum. This energy is referred to as short-wave radiation. We not only get warm from this energy, but it also illuminates the world for us.
The Earth and everything in it also emits radiant energy. Because the Earth is cooler than the sun, its characteristic radiation is not visible, but rather is in the lower energy infrared portion of the spectrum. This energy is referred to as long-wave radiation. The atmosphere not only receives this energy from the Earth’s surface, but the atmosphere, because it contains matter, radiates energy back downward to the Earth and as well as outward into space. Most of the time the atmosphere is at a cooler temperature than the Earth’s surface. Thus the amount of downward long-wave radiation to the Earth is less than the original long-wave radiation from the Earth to the atmosphere. The components leading to a net radiation balance are illustrated in Figure 2. The global Earth-ocean-atmosphere system is solely powered by the energy it receives from the sun. All processes taking place on Earth ultimately depend on this source of energy. After doing work, this energy stays in the system as heat. This heat could cause the temperature of the Earth system to continuously increase without limit if it were not removed. This catastrophe is prevented by the continuous loss of energy back to space by the long-wave radiation emission characteristic of all matter.

When the outgoing heat energy from the Earth exactly balances the incoming sun’s radiation, the Earth’s average temperature is stable. However since parts of the Earth system, such as the city of Houston, can exchange energy with other parts, these energy terms do not necessarily balance locally. Local temperatures are influenced by a complex array of other effects such as wind, evaporation and rain, and even heat from our furnaces. Heat energy is gained by some of these processes and lost by others. And the temperature we experience is the net result of the balance among them. As with all physical processes, energy must be conserved. The amount of energy entering a defined system (the city of Houston for example) during a particular time period such as a day, must equal the amount of energy leaving plus the amount of energy stored in the system during that time. This equation will allow us to look in a precise and quantitative way at the many differences between urban and rural environments. To do this we need to define several energy terms.

We have already become acquainted with the short and long-wave radiant energy terms as energy is transferred back and forth between the Earth and space. These we will specify as:

QI = Incident (direct and diffuse) solar radiation,
QR = Reflected solar radiation,
QLu = Upward surface emission of long-wave length radiation,
QLd = Downward atmospheric emission of long-wave length radiation,
Q* = Net solar energy (positive or zero) = QI - QR, and
QL = Net infra-red energy = QLd-QLu (generally negative or zero)

Other energy terms identify energy moving about the Earth and between the surface and the atmosphere. These types of energy are involved in modifying the temperature at various locations. They are:

QF = Anthropogenic energy such as from industry, transportation, heating and AC
QE = Latent heat from evaporation of water from trees, soil, bodies of water, etc.
QH = Sensible heat carried by vertical and horizontal air motion, including wind
QS = Storage heat flux within the system (ground, buildings, etc.).

There are many ways of expressing energy—calories, joules, British thermal units among others. The energy unit most commonly used to express the consumption of electricity is the kilowatt-hour (kWh). A kilowatt is equal to 1000 watts. Named in honor of James Watt the developer of the steam engine, a watt is really a unit of power and measures the rate at which energy is utilized. A watt is defined as one joule of energy per second. Therefore a kWh is equivalent to the consumption of 1000 joules every second for one hour. Since there are 3600 seconds in an hour, 1 kWh = 1000 x 3600 joules or 3.6 million joules or 3.6 megajoules (3.6 MJ). Much of our discussion here will be a comparison of the consumption of various energy forms over a period of one full day. Thus we may deal in units of kWh/day. And rather than working with total energy values for a whole ecosystem, we will focus on an individual square meter as a representative unit of area—thus energy values will be in units of kWh/day/m2 or MJ/day/m2. All of this nomenclature may seem rather complicated, but it is important to grasp it firmly before continuing to grapple with the bookkeeping of climate. Let’s try to think of it spending money rather than using energy. If one dollar is the monetary equivalent of one joule of energy, then spending one thousand dollars per second for an hour a day is the equivalent of utilizing a kilowatt-hour of energy during the day. This intensity of spending is equal to 3.6 million dollars a day just as the use of one kWh/day is equal to a daily energy usage of 3.6 million joules of energy or 3.6MJ/day. For the sake of comparison, 1 kWh or 3.6 MJ of energy is enough to vaporize (boil) about 1 2/3 quarts of water.

All of these energies vary from moment to moment during day and the year. And they also vary from one place to another. For example, the daily solar energy values for Houston presented in Figure 1 vary from one month to the next. For a given month a single value represents an average of all the instantaneous sunshine intensity values during the whole day for all days of the month over a 30 year period. Within a 24-hour day the solar energy varies, peaking sometime around noon and falling to zero between sunset and sunrise. Our task of understanding the urban heat island effect will be made much simpler by looking at time averaged energy values rather than constantly varying instantaneous values. An analogous situation is involved when planning one’s budget needs. A person need consider only monthly electric bills over the year rather than worry about minute by minute electric energy production requirements.

How are urban climate factors different from those of surrounding rural areas?

A diagram of the energy balance for a typical natural landscape in a rural location near Houston, Texas is shown in Figure 3a. A companion energy budget for an urban area such as Houston is shown in Figure 3b. The various energy terms vary from day to day, but those in the figure are characteristic of those for a typical sunny summer day in Houston and the surrounding rural environment.


Net energy from the sun and surface reflectance

Both rural and urban systems obtain energy from radiative processes, ultimately gaining energy from the sun and subsequently losing energy back to the upper atmosphere and space. Short-wave radiation from the sun is absorbed only during the daytime, but the long-wave radiation emitted by the earth system is lost all the time. Consequently, the earth’s surface warms during the day and cools at night by these radiative processes. The incident solar radiation, QI, is 7.6 kWh/m2/day for both locations because the same sun shines on both environments with equal intensity. With an albedo of 0.25 typical of a rural forest ecosystem, the reflected solar radiation, QR, is 1.9 kWh/m2/day in the country. However, because of a high density of low reflecting building and roadway surfaces in most cities, urban ecosystems may have an average albedo as low as 0.05, resulting in an urban reflected sunlight of only 0.4 kWh/m2/day. Thus the net energy obtained from the sun in the rural case is 5.7 kWh/m2/day while that in the urban case is 7.2 kWh/m2/day, or 1.5 kWh/m2/day more. As shown in Figure 1, an increase of 1.5 kWh/m2/day in solar radiation could result in a temperature increase of almost 13°F, changing April into June! Surface albedo is indeed a very important factor in urban heating and one that can be addressed by a change in the reflectivity of construction materials.

Radiant heat and temperature

Every material substance gives off radiant energy characteristic of the temperature of that substance. The long-wave upward radiation from both rural and urban environments will depend on the average effective temperature of the respective surfaces. Daytime rural summer surface temperatures in East Texas typically peak in the afternoon at about 30°C (86°F) with the nighttime minimum temperature a relatively cool temperature of 22°C (72°F). A realistic value of QLu for such a rural system is 8.5 kWh/m2/day. On the other hand, daytime urban surface temperatures can reach values well over 50°C (122°F) with a nighttime cooling of perhaps to 24°C (75°F). The value for QLd in the urban environment with this temperature range is 10.3 kWh/m2/day. The long-wavelength downward radiant energy back to the earth surface originates in the higher atmosphere when upward energy is absorbed, warming the atmosphere and causing it to reemit this energy. Downward atmospheric emissions of long-wavelength radiation are somewhat lower in energy than the companion upward radiation because the higher atmosphere, in general, is cooler than the earth’s surface and companion atmosphere. Because of close proximity, the effective temperatures of the higher atmospheres over both the rural and urban environments can be considered to be equal. High levels of pollutants in the urban atmosphere may be heat trapping and cause the urban upper atmosphere to be hotter than that of purer rural air. The presence of pollution would then cause the downward radiant energy term to be larger thus increasing urban heating. In this example the downward radiant energy flux is 5.9 kWh/m2/day. It is the same in both rural and urban environments, indicating that the temperature of the atmosphere is the same over both. The net loss of long-wave radiant energy from the rural environment is -2.6 kWh/m2/day and that from the hotter surfaces of the urban environment is -4.4 kWh/m2/day. The urban environment gains more energy from the sun than the surrounding rural area does, but, because of its higher temperature, it also loses more. Adding all four of the radiative energy terms together, we find that the rural area gains 3.1 kWh/m2/day while the urban area gains only 2.8 kWh/m2/day. If that were the whole story, the rural area appears to be gaining more energy per day than the urban area and thus should reach a higher average temperature. This is not the case because of other processes taking place and there is more to the story as can be seen by looking at the other energy terms in Figure 3.

Anthropogenic heat production

Everyone uses energy to make life more pleasurable and productive. How much residential energy is used depends on individual habits and needs, but is fairly accurately accounted for by one’s monthly fuel and electric bills. During the Houston summer, for example, about 75% of the residential use of electricity is for air conditioning. Likewise different people travel different distances in their cars with each person’s transportation energy reflected in what is spent for gasoline. Everyone who uses the products of or works in commercial businesses and industrial facilities ultimately pays for the energy used in these establishments.

A convenient way to keep track of all of the different direct human energy uses is to divide the total for the whole city by the number of citizens in the city. The result is the average per capita consumption of energy. This number is also the average per capita amount of energy added to the environment. For Texas, this number is about 460 kWhr/day/person. For comparison, all energy terms are given in kWhr/day/m2. To convert to these units, we need to know how many people occupy a square meter in Houston. According to the US Census Bureau (1990), Houston proper contains 1,631,000 people and covers an area of 540 square miles (1,382 square kilometers). A square kilometer contains a million square meters, so the area of Houston is 1,382,000,000 square meters. This results in a population density of 3,020 persons per square mile or 0.00116 persons per square meter. If one person produces 460 kWhr/day of energy, then one meter square’s worth of persons would produce 0.53 kWhr/day/m2. This figure compares well with the estimate of 0.5 kWhr/day/m2 depicted in the model shown in Figure 3. Note that no anthropogenic heat production is indicated in the rural energy balance. That is because the density of people in a rural environment is so low that the energy production of these people is negligible.

Latent heat and the vaporization of water

An energy term in Figure 3, which is very large in the country and much smaller in the city, is the latent heat term. This term has to do with the evapotranspiration of water. The latent heat is heat required in the vaporization of water. The process is called evapotranspiration because it is the combination of two processes, direct evaporation of standing and soil water and transpiration, which is the movement of water from the soil through plants into the atmosphere. The energy needed for both processes is the same. It is also the same as the heat energy needed to boil water on a stove. As energy is used to evaporate water from a surface, heat is withdrawn with the water vapor and the surface cools. This is true of our bodies as we cool by evaporating perspiration and it is true of the Earth as water vapor is formed and transported into the atmosphere. The latent heat of vaporization of water at normal ambient temperatures is approximately 2.46 million Joules per kilogram of water. Remember that one kWhr is 3.6 million Joules. Therefore a kilowatt-hour of energy will evaporate 3.6 divided by 2.46 or 1.46 kg of water. A kilogram is equivalent to 2.205 pounds.

We have been working with units of kilowatt-hours per meter square per day. Spread over 1 square meter, 1.46 kg of water would have a depth of 0.146 centimeters. So, if our rural environment spends 2.0 kWh/m2/day of energy on evaporating water, it will evaporate the equivalent of 0.292 centimeters of standing water per day. In one year or 365 days, that would amount to about 107 centimeters or 42 inches of standing water. Forty-two inches of water is essentially equal to the annual rainfall in the Houston area. Thus in our rural example, the latent heat loss indicates an assumption that the annual evapotranspiration is equal to the annual precipitation and that there is no water runoff or change in soil moisture content from year to year. If these assumptions are appropriate for the rural ecosystem, why is the urban system so different? The stated urban latent heat loss is 1.0 kWh/m2/day, 50% of the rural value, indicating that evapotranspiration in the city accounts for only half of the precipitation. Other processes must then remove the remaining water. A lack of trees and shrubs in the city reduces the amount of water that can be evaporated through transpiration, especially in industrial, commercial, and densely populated high-rise residential areas. In a rural situation with a healthy ground cover of grasses, shrubs and trees, perhaps half of the evaporative water loss is through the plants. Also, in a city, buildings, roads, and parking lots cover large portions of the surface. Over 50% of the ground area in central business districts is covered over and therefore sealed to water. This fraction lowers in more residential areas, but is probably of the order of 25%. These covered over and paved surfaces are impermeable to water and precipitation runs off into the storm sewer system rather than being available for evaporation. This impermeability not only affects the energy and heat balance of the city; it also contributes to other weather problems such as flooding.

Sensible heat: conduction, convection and advection

Sensible heat is heat that one can feel or sense. Sensible heat refers to the back and forth transfer of heat between the atmosphere and various surfaces. It is heat transferred by conduction within various surface materials and by convection (vertical) or advection (horizontal) wind as air moves from one place to another. The arrow denoting sensible heat in Figure 3 is bent to suggest both vertical and horizontal air transfers. It is rather difficult to specify a characteristic sensible heat value because it depends on so many variables. Generally, the atmosphere in contact with the surface of the Earth is warmer than the air above it because it receives heat from the warm surface. Vertical air motion arises when the warm surface air is lifted in the atmosphere by heavier colder air sinking. This vertical convection of air mixes the atmosphere within a certain boundary layer and is most noticeable in the morning as the sun begins the daily process of heating the Earth from the surface up.

Sensible heat associated with horizontal wind transfer depends on the speed, direction, temperature and humidity of the moving air. As an example, consider a dome of dry air 360 meters or 1000 ft. high extending over the city. A sensible heat transfer of 1.0 kWh/m2 from building hot surfaces to the air above them would warm the air dome by almost 14°F (7.7°C) through vertical mixing. The buildings would cool slightly because of the transfer of heat. This is how stored heat in structures such as buildings or room radiators heats up the air. Just as in the case of a room radiator, where the warmed air rises and mixes, carries off the heat from the source and spreads it throughout the room, the air over the city moves the heat from the source buildings and pavement. If the wind is blowing, this heated air then moves on to a new environment, leaving the city either somewhat cooler or warmer, depending on the temperature of the air moving in.

The prevailing wind in Houston is generally south to southeasterly which means that it blows in from the Gulf of Mexico, over the coast and up from Galveston or east of Galveston. The Gulf waters and associated air are usually cooler than Houston air. So one might expect that Gulf air blowing into Houston would replace its hotter air, leading to a negative contribution to the sensible heat term. That would be desirable, except this air is generally more humid than Houston air and thus inhibits local water evaporation and reduces any potential cooling by latent heat transfer.

Storage heat

Storage heat is energy that is absorbed by the various materials of the surface environment: buildings, pavement, soil, etc. During the day the energy from the sun heats different areas of the city to different degrees. For example the daytime summer temperature of an object such as an automobile sitting under the shade of a tree is noticeably cooler than one sitting in the full sun. The automobile in the sun has absorbed more storage heat. This storage heat will in large part be transferred to the atmosphere and the auto will cool in the evening after the sun sets. Some storage heat will remain in objects for several days and may not be completely released until a cooler day occurs.
Remember the interesting and related feature observed in the graph in Figure 1. We notice that the solar energy input starts to decrease in its annual cycle about a month before Houston’s average monthly temperature starts to fall. This "lag" in the temperature change indicates that heat is stored up somehow as the city warms up and then is slowly released later, warming the atmosphere even though the solar energy input is decreasing. The storage of the sun’s energy in buildings, streets and other surfaces is no different than the storage of energy from a cooking oven in food which remains hot for some time after it is removed from the oven and placed on the table. Just as the food does eventually become cool, so does the city begin to cool after the energy received from the sun starts its annual decrease, but the city may take several days or weeks to cool.

How can we vary the various energy factors to reduce urban heat?

Radiant energy and albedo changes

The main energy-inputs in any microclimate system are short-wave radiant energy from the sun and long-wave radiant energy exchange between the surface and the atmosphere. The only really effective way for humans to affect the radiation energy balance is by changing the albedo of the system so that more or less of the sun’s energy is reflected back into space. Whenever possible dark(low albedo) surfaces should be replaced by light(high albedo) surfaces so that less sunlight is adsorbed.

If there were no variation in these energy inputs, the temperature of the local environment would remain relatively constant for the whole year. That situation can be very close to what is observed in a tropical climate where the heat from the sun and other weather factors are very constant. However, climate seasons are evident even in the tropics. The sky is cloudier and the air temperature is cooler during the rainy season than during the dry season. In Houston, during the month of July, we have grown to expect little change in the daily temperature and humidity. But is that really what happens? Look at the data in Figure 4. Figure 4 shows the daily maximum temperature and the daily rainfall in Houston during August 2001. Although the temperature holds fairly constant during the first 26 days of the month, there are still some rather large changes. The 8°F fall in temperature from a high of 101°F on August 5 to 93°F a short two days later is probably due to the rain and clouds on August 6-7. Yet precipitation does not appear to be a factor in the 6°F one-day temperature drop and recovery on August 13 or the 8°F drop on August 19. On the other hand, a significant rain storm on August 26 and lasting until the end of the month correlates with a major temperature change of up to 20°F during the last week in August.

The point here is to note that all through the month of August, the sun’s energy at the top of the atmosphere changed very little, yet the day-to-day temperature changed dramatically. What caused this temperature change was the albedo change resulting from differences in cloud cover extent and thickness. Increased cloud cover during rainy days dramatically reduces the net amount of solar energy that penetrates into the earth-atmosphere system of Houston thereby reducing the maximum temperature of the city. Since the atmosphere itself does not significantly alter the short-wave energy from the sun, the albedo change due to clouds could be replaced by highly reflecting (high albedo) materials near the surface. These materials would decrease the net energy of the sun by reflecting it just as the clouds do and thus reduce the temperature of the surface. Such materials would be new light colored paving and roofing materials. In various parts of the city, paving and roofing materials cover as much as 50% of the surface and may have an average albedo as low as 0.10 to 0.15. If the rest of the surface is grass, trees, sand, and soil having an average albedo of 0.20- 0.25, the clear sky composite surface albedo would be about 0.15-0.20. If the roofing and paving material were replaced by highly reflecting materials with an average albedo of 0.65 to 0.75, then the average albedo of the city could increase as high as 0.40-0.50. This change would reduce the net solar energy absorbed by the city surface by almost 40% and result in a temperature change in the city very similar to that caused by a substantial cloud cover. Such a change could not only lower the temperature of the city to that of the surrounding rural temperatures, but could reduce it even further since the city albedo could be greater than the rural albedo.

Anthropogenic heat production change.

In an urban climate system radiant energies are joined by anthropogenic heat production. Since this energy is completely controlled by human activity, it is possible for man to heavily affect its influence on the local climate.

The energy balance example presented in Figure 3 shows an anthropogenic heat production of 0.4 kWhr/m2/day. This value is probably about right or perhaps a little low for Houston. The actual value is very difficult to estimate and will of course vary throughout the year as energy requirements change. It will also vary from one part of the city to another, being highest where there is a high density of power users, such as in the city center with many office buildings requiring lighting, heating, or air conditioning. It will also be relatively high in areas of concentrated heavy industry. Never the less there are certain general insights that we can obtain. Even though each person has different energy requirements, an average per capita energy usage is a valid way of assessing total usage in a city the size of Houston. One would not expect the Houston average energy usage per person to be the same as that for Chicago or New York because these places have different climates, energy requirements and infrastructure. For example, most people in Houston rely on their car a great deal whereas people living in Manhattan tend to use public transportation and taxis. Heating and cooling requirements are also different in the two cities.
Population density is another factor to consider in determining the anthropogenic heat production within a city. In Table 1, the 1990 national census data for the 20 most populated cities in the United States shows a broad range in population density.

Table 1: Anthropogenic heat production for the top 20 cities in the US. Table assumes each city has the national average per capita energy consumption of 11,230 Watts.

Land area (sq miles)
Persons per sq mile
Persons per sq meter
Anthropogenic heat (kWhr/m2day)
New York city, NY
Los Angeles city, CA
Chicago city, IL
Houston city, TX
Philadelphia city, PA
San Diego city, CA
Detroit city, MI
Dallas city, TX
Phoenix city, AZ
San Antonio city, TX
San Jose city, CA
Baltimore city, MD
Indianapolis city, IN
San Francisco city, CA
Jacksonville city, FL
Columbus city, OH
Milwaukee city, WI
Memphis city, TN
Washington city, DC
Boston city, MA


New York City has a density of 23,705 persons per square mile whereas Jacksonville, Florida has a density of only 837 persons per square mile. Assuming that everyone living in these cities consumes energy at the national average rate of 11,230 Watts, the anthropogenic heat production varies from a high in New York City of 2.30 kWhr/m2day to a low of 0.08 kWhr/m2day in Jacksonville, Florida. Other cities show energy values that range between these two. On the high end in such cities as New York where the population density is very high, this energy input is comparable to that of the sun! On the low end in cities where the population is really spread out, this energy input is essentially negligible. Houston falls on the low end of the scale, but not exactly negligible. It would probably not be productive to attempt to reduce the heat island effect in Houston by lowering the population density. On the other hand, we need to be aware that a lower population density may translate into additional heat generated by a greater use of transportation energy. As Houston grows, the population density should not be allowed to radically increase if we do not want the temperature to significantly increase with it.

As consumers, we can do many things to reduce the anthropogenic heat production in our city. One of the primary ways to do this is to modify our transportation behavior. We may reduce the amount of driving we do, own a more energy efficient vehicle, and use public transportation when ever possible. By making our energy usage more efficient, it is possible to reduce the anthropogenic heat production by 30-50%. By systematically reducing the population density of the city, this heat term can be made even smaller as can be determined by comparing cities of varying population density in Table 1.

Energy outputs as latent heat and sensible heat

Recall that the energy inputs of radiant energy and anthropogenic heat are balanced by energy terms such as latent heat, sensible heat and heat storage in the ground and other materials. Heat energy is generally carried away from the surface by latent heat.

Heat is also lost to the upper atmosphere by vertical sensible heat transfer as boundary layer air is vertically mixed. Lateral sensible heat is normally associated with winds. Winds can either heat or cool a region by replacing the air in the region with new air from a warmer or colder region up wind. The greater the sensible heat, the greater the cooling of the surface as energy is lost.

Latent heat is transferred from the surface to the upper atmosphere as water is vaporized at the surface either by simple evaporation or by transpiration through plants. The extent of evapotranspiration is dependent on the moisture content of the soil and other parts of the system and the ability of water to transport through the surface. A completely impermeable surface cannot evaporate water and a ground barren of trees and other plants cannot transpire. Increasing the number and density of trees and other plants will promote increased evapotranspiration. This will increase the latent heat energy lost from the surface and the surface temperature will decrease.

If the area is in the midst of a drought and the ground contains little or no water, evaporation will decrease. If, on the other hand, the lower atmosphere is saturated or near saturated with water vapor, evaporation will decrease. This situation is characteristic of Houston because of the presence of moisture-laden air from the Gulf of Mexico. In both cases, the latent heat term will be lowered, the surface system will tend to store energy as heat, and the temperature will increase.

The presence of a breeze will aid in the evaporation of water from the surface if the air moving into the region has a lower moisture content than the air it replaces. Thus, moving air will generally increase the amount of latent energy production, lowering the storage of heat and lowering the temperature. Taller buildings in the city center and satellite commercial areas tend to reduce the wind velocity near ground level unless the canyons between buildings (streets) are aligned with the wind direction. It is possible to orient and space such areas to optimize wind flow and thus aid in cooling that area of the city. The introduction of green spaces in large commercial areas will not only result in beautiful and functional park space, it can also aid in wind management, provide cooling shade and increase the amount of permeable soil and plants for increased cooling through latent heat transfer.

Houston is situated with respect to the Gulf of Mexico such that during the hottest times of the year the prevailing winds are from the southeast. As winds sweep in over the Gulf, they traverse an area that is composed of petrochemical industrial facilities and large areas of fallow fields with little moisture content. The winds gain a great deal of sensible heat from this area and its temperature rises. This hot wind then travels across Houston without bringing any comfort to the already hot city. If this area were to be developed so that its albedo were raised and planted with trees to increase evapotranspiration, a "peri-urban cool island" could be developed outside of Houston which would reduce the sensible heat load of down-wind Houston, thus cooling it. The peri-urban albedo could be raised in several ways. Trees alone would help a great deal. Highly reflective structures, commercial or residential could be put into place. It is even conceivable that the area could be developed into a large solar-power array that could be used to supply Houston with power that would replace much of the anthropogenic heat in the city. If the array were engineered out of a material base with a high albedo, then it would also aid in the formation of the "cool island." Dagobert Brito of the James A. Baker Institute for Public Policy at Rice University has estimated that a one hundred square mile array of solar panels would supply the power needs of all of Houston as well as form a high albedo area large enough to significantly cool the winds over Houston! A win-win situation that could eventually pay for itself. Perhaps other novel ideas could be developed that would be economically sound and climatically cooling.

How can we affect the quality of life by mitigating the Houston heat island?

To summarize, the various tools that can be employed to cool Houston are (1) planting trees in strategic locations; (2) using highly reflecting urban surfaces such as roofs, streets, and parking lots; (3) developing smart residential and commercial areas with appropriately placed green spaces; and (4) influencing the wind field with the development of cool islands upwind from the city. Together these techniques could reduce the heat load of the city by several kWhr/m2/day and reduce the temperature by as much as 10 or 15 degrees Fahrenheit. Cooling the city during the hottest days of the year will certainly increase our quality of life, simply be making it more comfortable to be outside. But there are other advantages to cooling the city.

The energy requirement and subsequent cost for air conditioning our living spaces increase directly proportional to the difference in temperature between the building or automobile interior and the outside. If the interior is kept at 75°F and the external temperature is reduced from 105°F to 95°F by heat island mitigation, the air conditioning bill can be reduced by 33%. This is a substantial economic saving as well as a reduction in the overall energy requirement of the city.

Houston has a pollution problem. A type of air pollution created from the combination of nitrogen oxides and volatile organic carbon compounds has been recognized to be particularly severe in Houston. This mixture in the presence of sunlight forms "photochemical smog." A major component of photochemical smog is ozone; other components are various peroxides and organic nitrogen compounds. Ozone can have adverse effects on humans, animals and plants. In humans it can cause decreased lung capacity, asthma, inflammation and swelling of lung tissue and other respiratory problems. Exposure to ozone can also impair one’s immune system defenses and cause susceptibility to colds, bronchitis and pneumonia.

Photochemical smog is visible as a haze that also contains small airborne particles or aerosols. The Houston area and several surrounding counties fail to meet the standards for ozone concentration set forth under the federal clean Air Act of 1970 and 1990 and administered by the Environmental Protection Agency. Ozone is known as the criteria pollutant for these standards. The Houston area is said to be in "nonattainment" for ozone and is required to take steps to lower the concentration to attainment levels.

Reactions that form photochemical smog develop when the sun is most intense, the air temperature is the hottest, the sky is cloudless, the wind is calm, and the concentrations of precursor pollutants are highest. These conditions are also those for which the heat island effect is strongest and mitigation efforts are most able to reduce the ambient temperature. Such reactions are sensitive to the ambient temperature and will slow down as the temperature is lowered. A rule of thumb in chemical reactions is that they will slow down by a factor of two when the temperature is reduced by 18°F. If heat island mitigation reduces the temperature on the hottest, windless days by as much as 7 to 8°F, the ozone concentrations may be reduced by as much as 25% and could be reduced from above to below the EPA mandated attainment level.


The temperature of urban areas can be markedly higher than that of the surrounding countryside. This fact is known as the "Urban Heat Island" effect. The interactions that account for this effect are many and they work together in a complex manner. Yet once we are able to recognize what these forces are and how they operate, it becomes possible to modify them in a way that reduces the temperature of the urban environment. These forces include:

Ways in which citizens can modify these forces to produce lower temperatures, obtain better quality of life, and foster enhanced public health are

The driving factors in the production of the urban heat island effect are many and are contributed to collectively by all city residents. The reduction of the urban heat island effect also must be a collective effort. Everyone in the city is a necessary participant in the process of reducing the urban heat island in Houston. All of us can help by educating ourselves on the subject, finding out what steps need to be taken by individuals and by groups, and convincing the city leaders, both political and economic, that they must truly be leaders for the creation of a cooler and more healthy Houston.

References for further study

Barry, Roger G. and Richard J. Chorley, Atmosphere, weather and climate, sixth edition, London and New York, Routledge, 1992.

Landsberg, Helmut E., The Urban Climate, New York, Academic Press, 1981.

Oke, T. R., Boundary Layer Climates, second edition, London and New York, Methuen, 1987.

The Heat Island Group, Environmental Energy Technology Division, Lawrence Berkeley National Laboratory, Berkeley, California,

Urban Climate Network, An Internet-based learning and data resource,

U. S. Environmental Protection Agency Global Warming Site, Local Actions,