2000: Microclimate Climate Change Case Studies

Problem

In 1995, the Intergovernmental Panel on Climate Change reported that since the 19th Century the Earth's global mean surface temperature has warmed between 0.3 and 0.6°C. Furthermore this report states, "The balance of evidence (from changes in global mean surface temperature and from changes in geographical, seasonal and vertical atmospheric temperature) suggests a discernable human influence on global climate." However, since many of our historic temperature measurements were made in urban regions, which are changing themselves, could these global temperature changes be due to local changes in the microclimates of growing cities? Understanding the extent of this influence is among the most important science problems facing climate researchers. However, whether these changes in temperature are local or global in origin, we are also interested to see how these changes may have caused changes in the frequency of extreme events (heat waves, cold periods, wet/dry period) in these cities. These two topics will be the focus of the Microclimate Team -- the connection between a city's population and the heat island effect, and the connection between changing microclimate and the frequency of extreme weather.

Objectives

Prepare four climate change case studies (each for a different city), investigating the science questions listed below to gather evidence about the extent to which human development may be influencing a particular urban area's climate. To make progress on this problem, each case study will analyze the relationships between long-term trends in population, the Urban Heat Island (UHI) Effect and the frequency of extreme climate events. The suggested urban areas represent different global regions: Case Study 1: Mexico City (Tropics), Case Study 2: New York City (Midlatitudes), Case Study 3: Moscow (High Latitudes) and Case Study 4: Hong Kong (Asia).

1. Science Questions for the Urban Heat Island Sub-Project:
   • How do the maximum (daytime) and minimum (nighttime) temperature trends compare between the urban station and the surrounding averaged rural area station data? Are they similar? Do the rates of change differ? How do the summer months compare to the winter months?
   • Is there evidence of the UHI signal?
   • How does maximum/minimum temperature change on decadal timescales for urban vs. rural area? Are the trends constant over the entire period?
   • What is the decadal rate of change for population in the urban area?
   • Do the results confirm the results of the University of British Columbia study that finds a relationship between increasing decadal changes in population and urban warming?
   • If there is a relationship, do the preliminary results confirm the population threshold found in the University of British Columbia study?
2. Science Questions for the Frequency of Extreme Climate Events Sub-Project
   • Are there long-term trends in the frequency of very cold (T-minimum), very hot (T-maximum) and/or rainy days/months for urban vs. rural areas? Is there an increase/decrease in climate extremes?
   • Are changes in extreme events observed only in certain months or seasons?
   • Are there decadal trends in the frequency of extreme climate events? Or is the trend constant over the entire record?
   • What is the decadal rate of change for population in the urban area?
   • How does the decadal trend in urban population relate to the decadal frequency of extreme climate events?
3. Science Question for the Joint UHI and Extreme Events Analysis
   • s there a relationship between trends in population, UHI and the frequency of extreme events?
   • f changes in extreme events are noticed, are these changes the same in the urban and rural regions?
   • s there a stronger UHI signal in areas with a certain population threshold?
   • ow do these urban case studies compare in terms of providing evidence of the influence human development has on regional climate change?
   • re climate change and population relationships detectable in some parts of the world and not others?
4. Building Team Members Skills and Knowledge. To gain some of the project background, the team will read and discuss the two student papers motivating this research. 1) University of British Columbia UHI Study and 2) the GISS/ICP study on the frequency of extreme climate events. (Day #1) Understand the major science concepts that:
   • There has been a trend in global warming since the industrial revolution. A significant portion of the science community is in agreement that human activities and natural variability both contribute to this climate change trend. (On-going discussion)
   • There are patterns of atmospheric circulation that produce different regional climates around the global. There also exist patterns of atmospheric circulation that can influence local climate. Therefore some regions may be warming less/more than others. (On-going discussion)
   • UHI is a phenomena associated with the development of intensely populated and industrialized global regions. Students will participate in a physical box model experiment developed by Umit Kenis to teach the concept that different surface areas have different radiative effects, absorbing and reflecting various amounts of the sun's energy. Removing vegetation also has a big effect on local temperatures. (July 12th)
   • If climate variables, such as temperature and precipitation vary significantly above or below the average for a particular time period, they often indicate extreme climate-related events. Such events include droughts, floods, severe storms, heat waves, blizzards and cold periods. (Internet Activity, July 13th)
   • Statistics that describe global or regional climate can change depending on the time and spatial scale you are investigating. (Excel Lab, July 11th)

Skill Development

Use Excel spreadsheets, conduct statistical analysis and graphing results. Students participate in a lab prepared by Umit Kenis where they compare daily temperature data taken during SI99 from a site on Broadway and another site in Riverside Park. They will create an Excel workbook and perform the following tasks: 1) set up a spreadsheet, 2) enter data, 3) graph trends, 4) Calculate the average and standard deviation, 5) plot correlations by using the best-fit line or logarithmic relationship, 6) Calculate the slope of the line or the rate of change.

Use the Internet as a research tool to obtain data and science background information. Umit and Leila will guide students in two Internet Challenges where they have to find and download data to answer specific science questions using the GISS and NOAA web sites.

Organize research data on the computer in formats conducive to preparing science presentations for progress reports, papers and final reports. Leila and Umit will work with the students to design procedures and formats for organizing the team data and results.

Tasks

One teacher-student team will be responsible for the UHI sub-project and the other will conduct the frequency of extreme events study. Each sub-project will produce analysis for targeted urban areas. New York, Mexico City, Moscow and Hong Kong are identified as the suggested cities because they represent different geographic latitudes or climate regions around the globe. Also, they are areas impacted by population growth.

Sub-project teams will compare their findings and report them as four-microclimate climate change case studies, including a discussion of what was collectively learned about the influence of human development on regional climate change. Are there trends in the data that provide evidence to either support or reject the connection between human activities and regional climate changes, extremes and/or variability?

Urban Heat Island Sub-Project. In this project we will assume that the size of urbanization is proportional to population. The main objective will be to plot the decadal change in maximum and minimum temperature (over the longest record possible) between urban weather stations and the average for surrounding rural stations and then correlate this trend to decadal trends in population.

1. Obtain minimum and maximum temperature data for urban and surrounding rural stations. This data can be found at the NOAA website.
2. Average minimum and maximum values for the 2-3 surrounding rural stations in each city region so that you have one mean value for the rural minimum temperature and one mean value for the rural maximum temperature.
3. Calculate the T-max and T-min differences between urban and the averaged rural stations for each city region at ten-year intervals.
4. Repeat this analysis to study seasonal differences, e.g. summer (JJA) vs. winter months (DJF).
5. Obtain the population data for as long a record as possible for each urban area. Be sure that the time series is similar to the climate data. This task should be coordinated with the Extreme Events sub-project since they will also be using these data. The following web sites can be used to access the data: US cities and cities outside the US. The NOAA web site used to obtain the temperature and precipitation data also has population data.
6. Plot population data for each city as a function of time. Add a line at ten-year intervals to study decadal trends.
7. Produce two plots correlating population trends to changes in summer/winter T-max and T-min differences. Is there a relationship between urban warming and population?
   • Summer: time series correlating differences between T-max and T-min between urban and averaged rural. Insert a best-fit line to show the correlation.
   • Winter: time series correlating differences between T-max and T-min between urban and averaged rural and decadal trends in population. Insert a best-fit line to show the correlation.

Frequency of Extreme Climate Events Sub-Project. For this project we will assume that an extreme climatic event is a record hot, cold, wet or dry month, that varies from the mean by greater than +1 or –1 standard deviation. The main objective is to identify any trends that may exist in the frequency of extreme events experienced in the regions of interest for the case studies and correlate these trends with trends in population.

1. For each case study city download the monthly mean T-maximum, T-minimum and precipitation data from the NOAA website. First check and make sure that the record provides at least 50 years of data. If it does not, you'll have to select another city for the case study.
2. Organize the data record in an Excel Workbook. Make separate files for summer (June, July and August) and winter (January, February and December) data records.
3. Calculate a summer and winter average (mean over the entire record) for each variable, T-min, T-max and precipitation. You should have six numbers to reflect the mean value of each variable for the two seasons.
4. Calculate the threshold that will define an extreme climate event. To determine these values use the statistical operation of standard deviation (SD). This defines variability from the mean, and the greater a particular monthly value is the value gets to +1 or -1 the more significant the change. ul>
5. Extremely cold month: When observed Tmin < Tmin(average)-Tmin(SD)
6. Extremely hot month: When observed Tmax > Tmax(average)+Tmax(SD)
7. Extremely dry month: When observed Rainfall < Rainfall(average)-Rainfall(SD)
8. Extremely wet month: When observed Rainfall > Rainfall(average)+Rainfall(SD)
9. Produce eight graphs listed below for each city to analyze trends:
   • Summer: Months per decade below the threshold calculated for T-min (maximum value allowed is 3 months &215; 10 years = 30)
   • Winter: Months per decade below the threshold calculated for T-min
   • Summer: Months per decade above the threshold calculated for T-max
   • Winter: Months per decade above the threshold calculated for T-max
   • Summer: Months per decade below the threshold calculated for average precipitation
   • Winter: Months per decade below the threshold calculated for average precipitation
   • Summer: Months per decade above the threshold calculated for average precipitation
   • Winter: Months per decade above the threshold calculated for average precipitation
10. Analyze decadal trends for temperature and precipitation. Add a best-fit line to each graph. You should have graphs for each season that represent record hot, cold, wet and dry months over time.
11. Calculate the slope of the line every ten years. Plot the numbers for the slope as a function of time. Does the slope change on a decadal scale?
12. Obtain the population data for as long a record as possible for each urban area. Be sure that the time series is similar to the climate data. This task should be coordinated with the UHI sub-project since they will also be using this data. The following web sites can be used to access the data: US cities and cities outside the US. The NOAA web site used to obtain the temperature and precipitation data also has population data.

Products

Four case studies presenting the joint results of the UHI and Extreme Events sub-project teams.

The Excel Workbook with all the data and results organized

Documented methods to conduct the UHI and Extreme Events studies. This document should be written so other students can replicate your methods during the academic year.

An outline prepared by the teachers describing the materials to include in a Research Project Guide for this team. The Guide will be used as a training resource for a teacher workshop designed to involve other educators and students in preparing similar case studies. This outline should identify skills and concepts to learn, existing lessons to include and/or develop, background materials, documentation needed to get the research started, produce results, present findings and evaluate research experiences.