Climatology Help (UHI)

You will need to use the rural station and urban station with values for each day. In urban environments the materials used have lower albedos, and thus absorb more radiation. Also, in urban environments precipitation does not have a chance to go into ground storage/sit at the surface and there is less vegetation, so evapotranspiration is low. This means excess solar energy is used for heating the surface instead of ET. Let’s look at the data and see if surface temperatures and humidities are warmer/drier in the urban areas.

1. Plot the temperature, at the urban station and the rural station (2 lines on plot). Connect the points from the urban station with a solid line and connect the points from the rural station with a dashed line.

Temperature at Urban and Rural locations: ( be sure to label the x and y axis as appropriate )

2. Compute the average temperature for your 8 observations at both the rural and urban stations.

3. Plot the relative humidity, at the urban station and the rural station (2 lines on plot). Connect the points from the urban station with a solid line and connect the points from the rural station with a dashed line.

Humidity at Urban and Rural locations: ( be sure to label the x and y axis as appropriate )

4. Compute the average relative humidity for your 8 observations at both the rural and urban stations

Statistical Significance

Now we will compute a Student’s T test in order to statistically determine the likelihood of these two sets of data being different.

•Our null hypothesis (or default guess) is “no difference” between the station data. This would

mean there is no Urban Heat Island detected.

•n = 8, the number of data points at each station.

Let’s look at the variation of both temperature and relative humidity between the two data sets.

5. Compute the standard deviation for temperature and RH at each station. Note: you can use the STDEV() function in excel. It would look like “=STDEV(B13:J13)”.

6. Next, compute the variance of the difference between the two means

7. And finally, compute the t value: (you computed averages already in #2 and #4)

(The equation below instructs you to take the absolute value of the difference between the two stations, then divide that by the variance that you calculated in part 6. This MUST be a positive number)

The degrees of freedom for this calculation is n+n+2, or 18 if n=8. If you were to look this up on a t value table (in the back of all statistic books), you would find that for 18 degrees of freedom the t value for probability of p=0.05 is 2.10. This means, that if your calculated t value exceeds the tabulated value (2.10), we can say that the means are significantly different at the 95% probability level and reject the null hypothesis of “no difference” between the stations.

8. Is your calculated t value less than or greater than the tabulated value? What can you say about the means of the two stations based on this t value?

temperature:

relative humidity: