Urban Hot Island Phenomenon of San Antonio Texas Using Time Series of Temperature From

“URBAN HOT ISLAND PHENOMENON OF SAN ANTONIO TEXAS USING TIME SERIES OF TEMPERATURE FROM MODIS IMAGES”

Sandra Ytuarte

Noreen O. Castellano

Hot Island Phenomenon Project Proposal

Urban Heat Islands have been created over time in the United States and around the world. Scientific data has shown that July’s maximum temperature during the last 30 to 80 years has been steadily increasing at a rate of one-half to one degree Fahrenheit every ten years.

The urban heat island phenomenon occurs when the air in the city is 2-8°F hotter than the surrounding countryside.[1] There are several factors that contribute to a city becoming a heat island. One scenario is when urban areas have fewer trees, and the lack of other natural vegetation to shade buildings. The vegetations purpose is to block solar radiation and cool the air by evapotranspiration. Evapotranspiration occurs when plants transpire water through pores in their leaves. The water draws heat as it evaporates, cooling the air. Another contributor of heat islands is the low reflectivity of roofs and paving materials. Buildings and pavements made of dark materials absorb the sun’s rays, causing the temperature of the surface and the air around them to rise. Pavements and dark materials such as a black surface in the sun can become up to 70°F hotter than a reflective white surface. The energy of the sunlight is converted into thermal energy and the pavement gets hot and causes the air to be heated around it. The pavement “Albedo” or reflectivity is the ratio of the amount of light reflected from a material to the amount of light shone on the material.[2] For pavements a lower albedo implies that more sunlight is absorbed by the pavement. As opposed to pavements with a higher albedo less sunlight is absorbed by the pavement and it is cooler. In addition, when the sun beats down on houses with dark shingle roofs, some of the heat collected by the roof is transferred inside which causes heat to be unnecessarily stored.

Smog is a serious effect created from an urban heat island. Smog is created when photochemical reactions occur due to pollutants in air that creates ozone in the lower atmosphere. The formation of smog is highly sensitive to temperature, the higher the temperature, the higher the formation and the concentration of the smog. These reactions are more likely to occur and intensify at higher temperatures.[3] Higher ambient temperatures in hot islands also increase the use of air conditioning. This increases the use of electrical power tremendously in large cities. As a power plant burns more fossil fuels, pollution levels, smog and energy costs are all increased.

Results Expected

Each city’s urban heat island varies based on the city structure and thus the range of temperatures within the island vary as well.[4]We expect that the downtown area of San Antonio will have a higher temperature than the hill country as a result from the absorption and storage of solar energy by the urban artificial substratum, and also from the heat released into the atmosphere from industrial and communal processes. In many large cities satellite sensed temperatures are 10-15°C warmer than the surrounding rural areas. [5]Another contributor to the increase in downtown temperature will be the lack of vegetation such as trees and shrubs. In summer, the symptoms of diurnal heating begin to appear by mid-morning and can be warmer than nearby downtown and the amount of less vegetation compared to the hill country.[6]

Methods

Data was inputted using NASA Earth Observing System Data Gateway. The test data was ordered through Land Processes DAAC (LP DAAC) User Services from the U.S. Geological Survey. The following parameters were needed: data set, data search type, area – latitude / longitude, date and time range. The following parameters were used: data set - MODIS Aqua (MOD 11-Land Surface Temperature & Emissivity), data search type – primary search, area -33 N 29S latitude -102 W -97E longitude, date range- July 1, 2004 through August 1, 2004, and time range - 2:00 p.m.

MODIS/Aqua (EOS PM) (MOD 11-Land Surface Temperature & Emissivity) satellites orbit around the earth and passes south to north over the equator in the afternoon (2pm). Aqua MODIS can view the entire earth’s surface every 1 to 2 days, acquiring data in 36 spectral bands, or groups of wavelengths.

An e-mail from CM Shared with the subject “ECS Notification” was received with five days. Within this e-mail was a URL link where you just clicked onto and the information was ready to be downloaded. This information was downloaded to the C drive of the computer. The MODIS Tool MRT was used the transfer the HDF files to tif files. See figure 1 on page 7 for MRT Tool. The input files (hdf) were opened one at a time. These file were give a new output file name (tif), output projection type: Geographic/WGS 84, projection parameter: of UTM Zone 14, and an output pixel size of 1000 m. Each tif file was opened in the ENVI software and these files were cut into sub sets for the San Antonio Area. The spatial data was set and these 30 tif files were stacked into one image file for night and day. The night and day image files were opened and only certain bands were selected for night and day. From this process four day and nine night were chosen after overlaying roads and creating a density slice for each. See figures 2-5 of July 13th on pages 8- 11 for the process used for selecting these bands for each night and day.

The four days that were chosen were July 4th, July 5th, July 13th, and July 18th. The nine night that were chosen were July 13th , July 14th , July 15th , July 16th , July 19th, July 20th , July 24th , July 27th , and July 31st. See the figure 6 on page 12 for 3-dimensional produced for further study.

Excel software was used to input formulas for converting DN numbers to degrees Kelvin. Degrees Kelvin was then converted to degrees Celsius and then finally to Fahrenheit. Tables were created to show the range of temperature results for each day and night. Average temperatures were then calculated for each day and night and graphed.

Results

Not all days or nights could be used to show the Hot Island Phenomenon. Some reasons for this were that there was simply no data available and the weather conditions were not favorable. Weather conditions that could have affected the MODIS/Aqua satellite are the following overcast or cloudy conditions, rain storms, and heavy humidity levels. The National Oceanic and Atmospheric Administration (NOAA) was used to locate weather data from July 2004. However, NOAA was not acquired due to charge for data requested.

The results from this data acquisition were not what we expected. The results that were observed from this data acquisition showed that July 13th had the hottest average day temperature of 118.4°F for the month of July 2004. July 10th had the hottest average night temperature of 85.2°F for the month of July 2004.

The July 16th and July 19th night images the showed the hottest temperature to be mostly in the downtown center of San Antonio. The July 4th, July 13th day images the showed the hottest temperature to be mostly in the downtown center of San Antonio. All the other images showed different areas that were hotter than the downtown center. Some of these areas are the following: Northwest - Medina Lake, Southeast - Lake Calaveras, Northeast - Canyon Lake, further out Northeast - Lake Travis, Northeast - where 410 and 35 intersect/ Interchange Parkway (Distribution Center), Northwest- IH 10 Interchange, Northwest - Medical Center Area, North - 281 airport area, and North - Canyon Springs Golf course.

Future Research

After one year using these mitigation strategies acquire new data using the same methods. Record and evaluate this new data to see if there has been any change in temperature for the San Antonio Urban Area. Data should be collected from NOAA for daily weather details and see if there is any correlation with the weather and the satellites ability to acquire data. This data will help improve our understanding of global dynamics of hot islands and the processes occurring in these urban areas. MODIS is playing a vital role in the development of validated, global, interactive Earth system models able to predict global change accurately enough to assist policy makers in making sound decisions concerning the protection of our environment. [7]

Conclusion

Mitigation strategies are needed to decrease the urban heat phenomenon. Some of theses strategies are the following: plant programs, installing reflective materials for roofs and pavements in urban areas.

Planting programs can help reduce the urban temperatures, for example one mature, properly watered shaded tree with a crown of 30 feet can “evapotranspire” up to 40 gallons of water in a day. Within ten to fifteen years a properly placed tree can grow to a useful size a can reduce heating and cooling costs by an average of 10-20%. Having trees in a city over period of time can be much less expensive than air conditioners and the energy needed to run them. To achieve the best results from the trees it is important to have a correct selection and location of the trees. Grouped trees can protect each other from the sun and wind, making them more likely to grow to maturity and live longer.[8] Deciduous trees shading the south and west sides of a building block the summer sun. Trees grouped together create a refreshing oasis in a city and also cool nearby neighborhoods. The leaves of a tree filter dangerous pollutants from the air. By introducing the use of canopy coverage of downtown sidewalk buildings, the temperature in the downtown areas can be reduced.

It has been discovered that buildings with light-colored roofs that reflect the sun’s rays use up to 40% less energy for cooling than buildings with darker roofs. A new rating system called the solar reflectance index (SRI) is being developed to measure how hot materials are in the sun. Traditional roofing materials have an SRI of 5% for brown shingles and 20% for green shingles.[9] The surface temperature of a roof is mainly determined by the vigorous heat flows at the outside surface. There are four different types of instruments used to characterize roofing samples: the Fourier-Transform Infra-Red (FTIR) Spectral Emissometer, the UV-VIS-NIR Spectrometer with an integrating sphere, the Solar Spectrum Reflectometer and the Emissometer.[10]The following three shingles are defined by their reflectance and temperature as the sun beats down on them. “Black Shingles” have a reflectance of 5% and can reach a temperature as high as 180°F, “Conventional White Shingles” have a reflectance of 29% and can reach a temperature as high as 157°F, and “Advanced White Shingles” have a reflectance of 60% and can reach a temperature of 128°F. [11]The advanced white shingle would lead to a cooler roof and, in turn, a cooler building. Light colored roofing materials have an increased albedo and therefore lower surface temperatures than dark colored materials.[12]

Increasing the albedo of buildings and pavement surfaces through the use of reflective paving materials will help to cool down the surrounding ambient air temperature. The reduction in temperature of the pavement due to a higher albedo would contribute to a reduced heat island effect. Dark fresh asphalt has an albedo of 0.05 and can reach a temperature of 123°F, a light aged asphalt has and albedo of 0.15 and can reach a temperature of 115°F, and an asphalt coated pavement has an albedo of 0.51 and a temperature of 88°F. More reflective pavements with higher albedos create a cooler atmosphere which would reduce the heat island effect.

Figure 1: MRT TOOL

Figure 2: July 13th Hottest Day Temperature

Without Road Overlay or Density Slice Range

Figure 3: July 13th Hottest Day Temperature

Road Map overlaid

Without Density Slice Range

Figure 4: July 13th Hottest Day Temperature

Density Slice Range

Figure 5: July 13th Hottest Day Temperature

Road Map overlaid

Density Slice Range Applied

Figure 6: 3 Dimensional - July 13th Hottest Day Temperature

References

M. G. Estes, Jr., V. Gorsevski, C. Russell, Dr. D. Quanttrochi, Dr. J Luvall, Approaching The Millenium, 1999 APA National Planning Conference, “The Urban Heat Island Phenomenon and Potential Mitigation Strategies”.

2 B. Pon, D. Moses Stamper-Kurn, C. Kenton Smith, and H. Akbari, “Existing Climate Data Sources and Their Use in Heat Island Research”, April 27, 2002.

3 M.Pomerantz, B. Pon, H. Akbari, and S.-C Chang. “The Effect of Pavements’ Temperatures On Air Temperatures in Large Cities” Heat Island Group, Lawerence Berkeley National Laboratory, Berkeley CA

4 Rosenberg, Matt T.. Urban Heat Islands http://geography.about.com

5 Price, J.C.. Monthly Weather Review 107, no. 11 (1999) pp 1554-1557. “Assessment of the urban heat island effect through the use of satellite data”.

6 Kim, H.H.. International Journal of Remote Sensing Vol 13, no.12 1992, pp 2319-2336. “Urban heat Island”.

7 M. Pomerantz, B. Pon, H. Akbari, and S. – C Chang. “Cool Roofs Instrumentation”, Heat Island Group, Lawrence Berkeley National Laboratory, Berkeley, CA

8 Ibid

9 Ibid

10 Ibid

11 A. Apprill, “Energy Emissivity Analysis of Georgia State University Building Rooftops”, p. 13, 1998.

12 Remote Sensing of Vegetation Lecture 9 http://ioc.unesco.org/oceanteacher/resourcekit/M3/data/datasets/earthdatapubs/Servers/MODIS.htm

2

[1] M. G. Estes, Jr., V. Gorsevski, C. Russell, Dr. D. Quanttrochi, Dr. J Luvall, Approaching The Millenium, 1999 APA National Planning Conference, “The Urban Heat Island Phenomenon and Potential Mitigation Strategies”.

[2] B. Pon, D. Moses Stamper-Kurn, C. Kenton Smith, and H. Akbari, “Existing Climate Data Sources and Their Use in Heat Island Research”, April 27, 2002.

[3] M.Pomerantz, B. Pon, H. Akbari, and S.-C Chang. “The Effect of Pavements’ Temperatures On Air Temperatures in Large Cities” Heat Island Group, Lawerence Berkeley National Laboratory, Berkeley CA