How does rising atmospheric CO2 affect marine organisms?

Click to locate material archived on our website by topic


Temperature - Urbanization Effects -- Summary
The warming of near-surface air over non-urban areas of the planet during the past one to two centuries is believed to have been less than 1°C.  Simultaneous warming in many growing cities, on the other hand, may have been a full order of magnitude greater.  Thus, since nearly all near-surface air temperature records of this period have been obtained from sensors located in population centers that have experienced significant growth, it is absolutely essential that urbanization-induced warming be removed from all original temperature records when attempting to accurately assess what has truly happened in the natural non-urban environment.

One popular method of minimizing the potentially huge spurious effect of urbanization on global or regional temperature trends is to only consider data from small towns; yet even modest congregations of people can generate heat of significant magnitude.  Oke (1973), for example, demonstrated that settlements with as few as 1,000 inhabitants typically exhibit heat islands on the order of 2 to 2.5°C - values that could be as much as four times greater than the mean increase in global air temperature that is believed to have occurred since the end of the Little Ice Age.

Changnon (1999) also emphasized this point in a study of soil temperatures at a rural site in Illinois that spanned the period 1889 to 1952.  His results revealed the existence of a significant urban-induced warming bias in nearby air temperature records that had not previously been detected, or even suspected.  Specifically, data from the rural setting revealed the existence of a temperature increase that was 0.17°C less than the 0.57°C warming determined from three benchmark stations with populations of less than 2000 people.  Changnon underscored the significance of his finding by stating that "this could be significant because the IPCC (1995) indicated that the global mean temperature increased 0.3°C from 1890 to 1950."

In addition to air and soil temperatures, water temperatures have sometimes been used to detect urbanization effects on nearby air temperatures over land.  Maul and Davis (2001), for example, analyzed air and seawater temperature data obtained over the past century at sites of primary tide gauges of the U.S. Coast and Geodetic Survey, calculating trends for the 14 longest records.  The mean century-long warming of these 14 sites was 0.74°C.  The result for Boston (a 100-year warming of 3.6°C), however, was almost seven times greater than the mean of the other 13 stations (0.52°C) and probably suffers from some unspecified error.

With respect to the results obtained for the 13 more consistent stations, 0.52°C of warming does not seem unusual for the 20th century, when the planet experienced the bulk of its recovery from the global chill of the Little Ice Age.  Nevertheless, the authors note that "each site has experienced significant population growth in the last 100 years, and ... with the increase in maritime traffic and discharge of wastewater one would expect water temperatures to rise," which implies that some unspecified portion of the half-degree C warming is due to a maritime analogue of the urban heat island.  And since Maul and Davis note that on the time scale they investigated "one would expect the water temperatures to equilibrate to the air," we are left with the conclusion that the true 20th century change in air temperature over the 13 U.S. coastal sites was likely significantly less than 0.52°C.

A study of 51 watersheds in the eastern United States also reveals the potential for an urban warming bias in climatic records that may previously have been thought to be unaffected by this phenomenon.  In this study, Dow and DeWalle (2000) report that a complete transformation from 100% rural to 100% urban characteristics results in a 31% decrease in watershed evaporation and a 13 W/m2 increase in sensible heating of the atmosphere.  Based upon their results, we have calculated that, to a first approximation, a transformation from a totally rural regime to a mere 2% urbanization regime could increase the near-surface air temperature by as much as a quarter of a degree Centigrade (See The Urbanization of America's Watersheds: Climatic Implications).  This powerful anthropogenic but non-greenhouse effect of urbanization on the energy balance of the watershed and the temperature of the boundary-layer air above it begins to express itself with the very first hint of urbanization and, hence, may be most difficult to remove from instrumental air temperature records that are used in attempts to identify any greenhouse warming that may be present.  Indeed, the warming effects of urbanization may already be present in many temperature records that have been considered "rural enough" to be devoid of all human influence, when such is really not the case.

Much the same thing has been observed at other places around the world.  In China, for example, Weng (2001) used remotely-sensed Landsat Thematic Mapper data in a Geographic Information System to determine the temperature consequences of the urban development of the third largest river delta in China (Zhujiang Delta) that followed economic reforms instituted there in 1978.  Between 1989 and 1997, it was determined that cropland area declined by almost 50%, while the area of urbanized land increased by nearly the same amount, with the result, in the words of Weng, that "urban development between 1989 and 1997 has given rise to an average increase of 13°C in surface radiant temperature."

In Australia, Torok et al. (2001) observed that urban-rural temperature differences scale linearly with the logarithms of city populations, as is true for cities in Europe and North America.  They also learned that Australian city heat islands are generally of lesser magnitude than those of similar-size European cities, which are typically of lesser magnitude than those of similar-size North American cities.  The regression lines for all three continents, however, converge in the vicinity of a population of 1000 people, where the urban-rural temperature difference is approximately 2.2 ± 0.2°C, essentially the same as what Oke (1973) reported two decades earlier.

In addition to city temperatures increasing in response to increases in city population, it is possible for urban-rural temperature differences to intensify in cities experiencing no change in population, as demonstrated by Bohm (1998), who analyzed urban, suburban and rural temperature records in and around Vienna, Austria over the 45-year period 1951 to 1996.  During this time, Vienna experienced zero population growth.  However, there was a 20% decrease in woodland area and a 30% decrease in grassland area within the city, as well as a doubling of the number of buildings, a ten-fold increase in the number of cars, a 60% increase in street, pavement and parking area, and a 2.5-fold increase in energy consumption.  As a consequence of these changes, suburban stations exhibited temperature increases ranging from 0.11 to 0.21°C over the 45-year period, while urban stations experienced temperature increases ranging from zero, in the historic center of the city, to 0.6°C in the area of most intensive urban development.

In light of these several findings, there is clearly ample opportunity for large errors to occur in attempts to reconstruct non-urban temperature trends of the past century or so.  Given the magnitudes of these potential errors, which often rival or vastly exceed the magnitude of the purported non-urban global temperature trend, it appears that more detailed analyses of urban population and development characteristics are needed before we can be confident that the global temperature record of the past century or so is properly corrected for these phenomena.  And until this is done, it would be premature to put too much faith in that record as it stands today.

A case in point is provided by Hasanean (2001), who investigated near-surface air temperature trends in eight Eastern Mediterranean cities.  All of the stations with sufficiently long records exhibited similar uniform warming trends that began about 1910; but only some of them exhibited less coherent and discontinuous warming trends in the 1970s.  In view of what we know about urban effects on temperature, the first of these warming trends may well represent, at least partially, a true background warming of the entire region; but the divergent temperature histories of the cities in the 1970s and beyond may well be expressions of differences in their rates of urban growth and development.  Of the eight cities studied, in fact, four of them exhibited overall warming trends while four of them exhibited overall cooling trends, allowing one to say little about the mean temperature trend of the entire region.

A study from Norway raises similar concerns.  Nordli (2001) developed a number of summer temperature histories for rural areas of the central and southern parts of the country based on relationships established between modern instrumental temperature records and dates of local grain harvests of barley and oats, finding that summer temperatures during the last part of the Little Ice Age (1813-1880) were about 1°C lower than those of the last 70 years, while "the warmest decade of the series is the 1930s."  This rural-based assessment of climate change suggests there has been no net warming of the Norwegian countryside for the past seven decades, in contrast to what is claimed for the world by the Intergovenmental Panel on Climate Change on the basis of almost exclusively urban records.

Holmgren et al. (2001) have observed much the same thing in South Africa.  Based on a 3000-year temperature history for the southern part of the continent - which they derived from a correlation between color variations in the annual growth layers of a stalagmite and an area-averaged regional temperature series - they were able to clearly delineate the existence of both the Medieval Warm Period and Little Ice Age.  As for the late nineteenth century warming that appears in the instrumental surface record, however, they could find "little evidence" of it in the stalagmite data, which are not affected by the phenomena that produce the urban warming that typically contaminates near-surface air temperature records of even small settlements.

With respect to this latter topic, i.e., causes of urban heat islands, Balling et al. (2002) recently investigated the role played by the buildup of carbon dioxide in the air over cities that results from the copious quantities of CO2 emitted to the atmosphere by vehicular traffic and industrial processes (see Urban CO2 Dome in our Subject Index).  Using a one-dimensional infrared radiation simulation model supplied with data on atmospheric CO2 concentration, temperature and humidity obtained from vertical profiles of these parameters measured during daily aircraft flights that extended through and far above the urban CO2 dome of Phoenix, Arizona, they calculated that the excess CO2 over the city produced by the burning of fossil fuels creates a maximum warming that is one to two orders of magnitude less than that produced by the other urban heat island-producing phenomena of the city.

What is the bottom line with respect to all of these observations?  Hegerl et al. (2001) indicate that all instrumental long-term near-surface air temperature data sets are plagued by systematic errors that "cannot be assessed at present," one of which they say is "urban warming."  In this assessment they are correct.  Nevertheless, they go on to do precisely what they say cannot be done, claiming that "the effect of urbanization on estimates of global-scale signals should be small."  From the several reports discussed in this brief summary, however, it is clear that nothing could be further from the truth; the effects of urbanization are typically far larger than the background signal being sought.  Therefore, since the true temperature history of the non-urbanized area of the planet over the past couple of centuries is so much smaller than the warming produced by growing urban heat islands over this period, it must be admitted that the contemporary natural temperature trend of the planet cannot be accurately determined from the near-surface air temperature record.

References
Balling Jr., R.C., Cerveny, R.S. and Idso, C.D.  2002.  Does the urban CO2 dome of Phoenix, Arizona contribute to its heat island?  Geophysical Research Letters 28: 4599-4601.

Bohm, R.  1998.  Urban bias in temperature time series - A case study for the city of Vienna, Austria.  Climatic Change 38: 113-128.

Changnon, S.A.  1999.  A rare long record of deep soil temperatures defines temporal temperature changes and an urban heat island.  Climatic Change 42: 531-538.

Dow, C.L. and DeWalle, D.R.  2000.  Trends in evaporation and Bowen ratio on urbanizing watersheds in eastern United States.  Water Resources Research 36: 1835-1843.

Hasanean, H.M.  2001.  Fluctuations of surface air temperature in the Eastern Mediterranean.  Theoretical and Applied Climatology 68: 75-87.

Hegerl, G.C., Jones, P.D. and Barnett, T.P.  2001.  Effect of observational sampling error on the detection of anthropogenic climate change.  Journal of Climate 14: 198-207.

Holmgren, K., Tyson, P.D., Moberg, A. and Svanered, O.  2001.  A preliminary 3000-year regional temperature reconstruction for South Africa.  South African Journal of Science 97: 49-51.

Maul, G.A. and Davis, A.M.  2001.  Seawater temperature trends at USA tide gauge sites.  Geophysical Research Letters 28: 3935-3937.

Nordli, P.O.  2001.  Reconstruction of nineteenth century summer temperatures in Norway by proxy data from farmers' diaries.  Climatic Change 48: 201-218.

Oke, T.R.  1973.  City size and the urban heat island.  Atmospheric Environment 7: 769-779.

Torok, S.J., Morris, C.J.G., Skinner, C. and Plummer, N.  2001.  Urban heat island features of southeast Australian towns.  Australian Meteorological Magazine 50: 1-13.

Weng, Q.  2001.  A remote sensing-GIS evaluation of urban expansion and its impact on surface temperature in the Zhujiang Delta, China.  International Journal of Remote Sensing 22: 1999-2014.