East Rutherford, NJ February 2, 2014

Why would I, or anyone for that matter, care what the temperature for East Rutherford, NJ will be on a specific date? Normally, no one would care other than those who live there, but the answer is simple. East Rutherford, home to MetLife Stadium, has been selected to host Super Bowl XLVIII on Sunday, February 2, 2014. Super Bowl, considered by many to be an unofficial American national holiday, is one of the most-watched sporting events in the world. Continuing to gain popularity not only for the game itself, but also for the absurdly expensive and extremely entertaining commercials and the half-time performance; Super Bowl XVIII is expected to have an economic impact of approximately $500 million. So why did I choose to do my report on Super Bowl 2014? Well, I am a big fan of the Super Bowl and all I’ve been hearing lately is The Farmer’s Almanac prediction of a bitterly-cold, wet and white start to February for the mid-Atlantic region of the United States (Almanac Publishing Co.). So I thought it would be interesting to take a look into what type of weather the teams and performers would be challenged with.
Image from 2013 Farmer’s Almanac with regional predictions for central and eastern US.

For this reason, the focus of my research paper will be an analysis of variables directly affecting weather for East Rutherford, NJ on February 2nd, over that past 30 years. The dependent variable, wind chill, is determined by independent variables: wind speed, temperature, humidity and whether or not Punxsutawney Phil saw his shadow (otherwise known as could cover). The most important independent variable in determining wind chill during the winter months is temperature. The Wind Chill is the temperature your body feels when the air temperature is combined with the wind speed. My dependent variable is the wind chill for East Rutherford, NJ on February 2nd, over that past 30 years. Wind chill is defined by The Free Dictionary is as “the serious chilling effect of wind and low temperature: measured on a scale that runs from hot to fatal to life and allows for varying combinations of air temperature and wind speed”. Wind chill is a crucial factor in winter weather and can be used to determine various aspects from proper outdoor dress attire to limitations on the human body when exposed to the elements. My primary independent variable is temperature. Temperature, defined as a measure or degree of hotness or coldness, is a basic element of weather and climate (Farlex, Inc.). While there are many scales in which temperature is measured, I will be using degrees Fahrenheit (º F) for my data, as this is the commonly used scale in the United States. Simply knowing the temperature doesn't convey enough detail about weather conditions for you to dress wisely for winter weather; however, when calculated with wind speed this can provide a relatively accurate projection of the conditions on a cold day.

My second independent variable is wind speed. Wind speed is defined as the moving of air, especially a natural and perceptible movement of air parallel to or along the ground (Farlex, Inc.). As wind speed increases and temperature decreases this causes the wind chill to worsen, with a possibility of resulting in what is known as a ‘wind chill advisory’. My third independent variable is humidity. For the purposes of my report, I will be placing my focus on relative humidity. Relative humidity, as defined by the Free Dictionary, is a ratio of the amount of water vapor in the air at a specific temperature to the maximum amount the air could hold at that temperature, expressed as a percentage. In colder weather, high humidity makes for penetrating cold due to heat transfer from the body to surrounding air, thus contributing to a colder perceived temperature. When you’re outside on a cold day, does it feel better to stand in the sun or in the shade? If we approach this with the same mentality as Punxsutawney Phil, the groundhog who sees his shadow and runs, then we’ve clearly got something backwards. Cloud cover has a direct influence on perception of outside temperatures due to absorption of electromagnetic radiation into the skin. When rays are absorbed into the skin we feel warmer; now take that analogy and apply it to heating the surface of the earth, because of this sunshine has the ability to increase the wind chill temperature by 10º to 18º F (National Oceanic and Atmospheric Administration). The majority of the data used in this report was found through weather archive sites such as Weather Underground, the U.S. Department of Commerce’s National Oceanic & Atmospheric Administration, and the National Weather Service. The Weather Underground and National Weather Service sites were primarily used to locate the wind speed, the temperature, humidity percentage and cloud cover details on February 2nd of each year dating back to 1981. The CDC’s NOAA website was used to obtain a chart on how to calculate wind chill.

East Rutherford, NJ Dataset
Test Data Scale
(days)

Dependent Wind Chill
(º F)

Primary Independent Wind Speed
(mph)

Independent Temperature in East Rutherford
(º F)

Independent Humidity 67 60 62 83 Independent Humidity 61 61 66 73 72 68 66 68 43 71 63 92 73 74 60 43 54 41 43 54 80 76 96 79 81 89 72 85 75 81

Independent Cloud Cover
1 = Shadow 2 = No Shadow

2/2/2014* 2/2/2013 2/2/2012 2/2/2011 Research Data Scale
(days)

28 14 42 32

8 9 7 7 Primary Independent

34 24 46 37 Independent Temperature in East Rutherford
(º F)

1 2 1 2 Independent Cloud Cover
1 = Shadow 2 = No Shadow

Wind Chill
(º F)

Wind Speed
(mph)

2/2/2010 2/2/2009 2/2/2008 2/2/2007 2/2/2006 2/2/2005 2/2/2004 2/2/2003 2/2/2002 2/2/2001 2/2/2000 2/2/1999 2/2/1998 2/2/1997 2/2/1996 2/2/1995 2/2/1994 2/2/1993 2/2/1992 2/2/1991 2/2/1990 2/2/1989 2/2/1988 2/2/1987 2/2/1986 2/2/1985 2/2/1984 2/2/1983 2/2/1982 2/2/1981

27 40 28 34 37 31 22 33 20 31 12 41 35 36 17 22 11 2 18 33 42 39 42 36 38 20 18 39 21 33

3 3 12 4 4 1 5 13 13 6 14 5 3 6 6 12 3 14 15 6 7 9 8 5 8 12 5 9 10 19

30 42 36 37 40 32 28 40 30 36 24 44 37 40 24 31 16 16 29 38 46 44 46 40 43 30 24 44 30 42

1 1 1 2 1 1 1 1 1 1 1 2 1 2 1 2 1 1 1 1 2 1 2 1 2 1 1 2 1 1

Wind Chill compared to Temperature in East Rutherford 45 40 35 Wind Chill (º F) 30 25 20 15 10 5 0 10 20 30 40 Temperature in East Rutherford (º F) 50 y = 1.207 x - 13.217 R² = 0.933

    .

The slope of the regression line is positive. The R2 of .933 means 93.3% of the variation in wind chill is accounted by variation in temperature. The R value of .966 shows a strong positive correlation. T-value = 19.7531; when calculated at 95% confidence level, the null hypothesis of ‘no correlation’ is rejected, suggesting there is correlation between temperature and wind chill.

Wind Chill compared to Wind Speed 45 40 35 Wind Chill (º F) 30 25 20 15 10 5 0 0 5 10 Wind Speed (mph) 15 20 y = -0.784 x + 34.876 R² = 0.111

   

The slope of the regression line is negative. The R2 of .111 means 11.1% of the variation in wind chill is accounted for by variation in wind speed. The R value of .3332 shows a moderate inverse correlation. T-value = 1.87; when calculated at 95% confidence level, the null hypothesis of ‘no correlation’ is accepted, suggesting the correlation may be a chance based on the sample.

Wind Chill compared to Humidity 45 40 35 Wind Chill (º F) 30 25 20 15 10 5 0 0 20 40 60 Humidity 80 100 120 y = 0.473 x - 3.898 R² = 0.426

   

The slope of the regression line is positive The R2 of .426 means 42.6% of the variation in wind chill is accounted for by the variation in humidity. The R value of .6527 shows a strong correlation. T-value = 4.56; when calculated at 95% confidence level, the null hypothesis of ‘no correlation’ is rejected, suggesting there is correlation between humidity and wind chill.

Wind Chill compared to Cloud Cover 45 40 35 Wind Chill (º F) 30 25 20 15 10 5 0 0.5 1 1.5 Cloud Cover
(1 = Shadow 2 = No Shadow)

y = 11.114 x + 14.523 R² = 0.225

2

2.5

   

The slope of the regression line is positive The R2 of .225 means 22.5% of the variation in wind chill is accounted for by variation in cloud cover. The R value of .4743 shows a moderate correlation. T-value = 2.85; when calculated at 95% confidence level, the null hypothesis of ‘no correlation’ is rejected, suggesting there is correlation between cloud cover and wind chill.

The Durbin Watson analysis provided a value of 2.30, meaning autocorrelation is inconclusive; regression statistics may or may not be biased or unreliable. The regression may or may not be used to determine the true relationship between the variables.      Less than dL (< 1.14 ) = Positive autocorrelation Between dL and dU ( 1.14 and 1.74 ) = Test not conclusive ( Don’t Know ) Between dU and 4dU ( 1.74 and 2.26 ) = No autocorrelation Between 4dU and 4dL (2.26 and 2.86 ) = Test not conclusive ( Don’t Know ) Greater than 4dU ( > 2.86 ) = Negative autocorrelation

Using Durbin Watson, I was able to obtain the x-values needed to process the Multiple Linear Regression and Correlation Analysis. Below I have analyzed the 2013 and 2014 predictions to determine if the wind chill meets or is colder than expected.

Estimation Statement: = -7.4100+ (-.5907) X 1 + (-.5907) X 2 + (1.1885) X 3 + (-.3097) X 4 Regression output (2013) Variables Intercept Wind Speed Temperature in E. Rutherford Humidity Cloud Cover

coefficients -7.4100 -0.5907 1.1885 -0.00049033 -0.3097

X -7.41

A -0.5907 1.1885 -0.00049033 -0.3097

B 9 24 60 2

Totals -7.41 -5.317 28.524 -0.029 -0.619 15.15

2013 Analysis: The estimated wind chill for 02/02/13 was 15º F. The actual wind chill was 14º F; meaning the wind chill was colder than anticipated.

Regression output (2014) Variables Intercept Wind Speed Temperature in E. Rutherford Humidity Cloud Cover

coefficients -7.4100 -0.5907 1.1885 -0.00049033 -0.3097

X -7.41

A -0.5907 1.1885 -0.00049033 -0.3097

B 8 34 67 1

Totals -7.41 -4.726 40.409 -0.033 -0.310 27.93

2014 Analysis: The estimated wind chill for 02/02/14 is 28º F. If the calculation works correctly, the actual wind chill should fall approximately 1-2º F above that estimation. The primary purpose of this paper was to determine the estimated wind chill for East Rutherford, NJ on February 2, 2014. This has been completed by analyzing the relationship between 5 variables (1 dependent; 4 independent); however the results were inconclusive due to the likely hood of many other factors influencing how cold or warm it will be in a certain region on a specified date.

Works Cited
Almanac Publishing Co. The Farmers' Almanac. Ed. Peter Geiger and Sondra Duncan. Almanac Publishing Co., 2013. Farlex, Inc. The Free Dictionary. 2013. September 2013 . National Oceanic and Atmospheric Administration. National Climatic Data Center. Ed. US Dept of Commerce. 2013. September 2013 . National Weather Service. Wind Chill Chart. 2012. 2013 September . Stormfax, Inc. Stormfax Weather Almanac - Groundhog Day. 1996-2014. September 2013 . Weather Underground, Inc. Wunderground. 2013. Septermber 2013 .