Abstract

 The effects of climate change on weather patterns is a long-discussed topic in the scientific community.  Researchers have created simulations, models, and predictions in an attempt to explain future weather patterns such as hurricanes. The purpose of this paper is to validate whether some of these predictions are currently observable; for instance, as increases in hurricane wind speeds and the increase in the frequency of high category hurricanes.  An experiment was created whereby t-tests were used on hurricane data between 1970 and 2015.  The experiment was designed to determine if there were significant changes during the 45 years.  Additionally, the hurricane data was analyzed by calculating the correlation coefficients between each data type and the sea surface temperature (SST) to ascertain whether there was a statistically significant difference (e.g. change) between them.  The results of the experiment showed that there was an increase in average wind speeds for hurricanes over the past four and a half decades, as well as, a decrease in the frequency of tropical storms and tropical depressions. Furthermore, there were moderate correlations found between the individual hurricane data items and SST anomalies, indicating that there are other factors affecting hurricane patterns.  It is recommended that future studies attempt to eliminate the limitations associated with this research and study how varying UV radiation from the sun may be linked to hurricane patterns.

 Keywords:  Climate Change, Hurricanes, Sea Surface Temperature

Hurricanes and Climate Change

Identifying a Relationship between the Two Occurrences

Introduction

Climate Change as a global issue

 The issue of climate change remains an ever-growing concern, outside the academic world.               Many world leaders have attempted to address the issue of climate change in the past through agreements, such as the Kyoto Protocol to the United Nations Framework Convention on Climate Change (1997); however, they have yet to make real progress in halting the rising temperatures. The more recent Paris Climate Agreement made greater attempts to handle the rising temperatures; however, the agreement still does not limit the rising temperatures enough to halt its effects.

Get Help With Your Essay
If you need assistance with writing your essay, our professional essay writing service is here to help!
Essay Writing Service

Climate Change and its effects on Weather

 Climate change discussion has affected the public in many ways—from increased awareness to individuals, championing change in government policy.  Studies showing that the polar ice caps are melting validate the predictions of Global Warming; however, it is currently difficult for the public to identify how climate change is affecting the weather.  Usually, after a natural disaster, many news publications write stories, questioning whether climate change was a cause of the disaster.  Some scientists have even argued that the recent drought and subsequent forest fires in California were more severe because of climate change (Bedsworth, Cayan, Franco, Fisher, Ziaja, 2018).  Due to the intricate relationship between hurricanes and temperature in the ocean, it is believed that climate change and hurricanes have a relationship.

Climate Change and recent Hurricanes

 With devastating hurricanes such as Sandy, Maria, and Harvey all occurring within the last decade, it is believed by many that there is a relationship between Climate Change and the intensity, as well as, the frequency of hurricanes.  Hurricanes, according to the National Oceanic and Atmospheric Administration (NOAA), are low-pressure tropical cyclones that require warm water to begin their formation; as well as, their ability to maintain their strength (2013).  The effect of Climate Change on Sea Surface Temperature (SST) is believed to be the area that links Climate Change and hurricane intensity and frequency.  It is best to examine the current and predicted findings on the relationship between climate change and hurricanes to identify how climate change affects hurricane patterns

Literary Review

Links Between Climate Change and Hurricanes

 Before covering the potential linkage between hurricane intensity and rising temperatures, it is necessary to see how the frequency of hurricanes may change due to climate change.  According to the Geophysical Fluid Dynamics Laboratory (2018), despite the frequency of tropical cyclones in the Atlantic showing an increase in frequency based on historical data, adjustments due to historical inaccuracies clearly demonstrate that there are no apparent links between Climate Change and the frequency of tropical cyclones.  There is; however, a consensus under the U.S. Global Change Research Program that since the 1980s, there has been an apparent increase in the frequency of category 4 and 5 hurricanes (Walsh et al., 2014).  Furthermore, current models predict a significant increase of high-category hurricanes; as well as, a subsequent decrease in tropical storm frequency through the 21st century (Knutson et al., 2013).  These predicted changes lend researchers to accept that Climate Change does affect the intensity of hurricanes and changes their frequency of occurrence.  While it is extremely likely that these effects are a result of the increasing SST, it must be considered, however, that there are other areas climate change that could be causing the effect—especially when considering tropical cyclone growth.

Other Climate Change Factors

 The attempt to link climate change and hurricane patterns thus far has only compared sea surface temperature to hurricanes.  There are alternatives factors that must be considered when identifying the relationship.  For instance, based on a statistical analysis conducted by Elsner & Jagger (2009), they determined that solar activity of the sun has an inverse relationship with tropical cyclone frequency, as well as, their intensity. They explain that:

… an increase in solar UV radiation during periods of strong solar activity will have a negative influence on tropical cyclone intensity as the temperature near the tropopause will warm through absorption of the radiation… (p. 65)

Thus, it must be understood that while climate change may be a factor in hurricane intensity, the UV radiation from the sun, must be factored into hurricane patterns.

Present Day Validation of Models

 When considering the current predictions of future hurricane patterns as a result of climate change, it is necessary to understand that hurricane detection mechanisms were not as sophisticated as present-day technology.  As a result, it is probable that the hurricane data used by models, including 19th century, as well as, early 20th century data are lacking enough information to provide an accurate prediction of future hurricane patterns.  Thus, this paper will attempt to use recent data, between the years 1970 and 2015, to identify if the increase in SST anomalies has resulted in the frequency changes in hurricanes and, subsequently an increase in wind speeds.

Methodology

 Questions and Hypotheses

 Identifying whether Climate Change has effects on hurricane patterns that are observable in the present requires designing the experiment with several hypotheses. A series of questions and hypotheses were created to identify how hurricane data needs to be tested.

 Q1. Is there a statistically significant change in yearly average hurricane data between 1970 and 2015?

 H10. There is no statistically significant change in yearly average hurricane data between 1970 and 2015.

 H1aThere exists yearly average hurricane data that experienced statistically significant change between 1970 and 2015.

 Q2. If there exists yearly average hurricane data that experienced a statistically significant change between 1970 and 2015. Does a correlation exist between that data and rising SST anomalies?

 H20.  There is no correlation between the data that experienced statistically significant change and rising SST anomalies.

 H2aThere exists a correlation between the data that experienced statistically significant change and rising SST anomalies.

 Defining Variables of Experiment

 The variables that will be used to answer Q1 and Q2 are

The yearly average frequency of high category hurricanes (4 & 5);

The yearly average frequency of tropical storms and depressions

The yearly average frequency of all tropical cyclones (including hurricanes and tropical storms)

The yearly average frequency of hurricanes of categories 1,2, and 3

The yearly average hurricane pressure in millibars (mb)

The yearly average hurricane wind speed (knots)

The purpose of testing the yearly average frequency of tropical cyclones of varying intensity was to identify how the frequency of varying hurricane types might have changed between 1970 and 2015. It was mentioned before, that current models predicted high category hurricanes to experience an increase in frequency in the future, thus it was necessary to test high category hurricanes for Q1 and Q2. The same applies to the other yearly average frequencies because should the predictions be correct, their frequency should be affected. The yearly average hurricane pressure was to recognize whether the Sea Surface Temperature anomalies had an effect on the pressure, as it may have an effect on hurricane intensity. The yearly average wind speed was tested in the event that H10was true for high category hurricane frequency. This would demonstrate whether the intensity of hurricanes still increased as it was possible that wind speed did increase but not enough for hurricanes to be considered. Each variable would be used to answer Q2, should H1abe true that particular variable. With the variables identified, the next step required the collection of data, to perform the statistical tests.

Data Collection

 The purpose of this paper is to identify whether there is an observable change predicted by the computer models and to identify if those changes are the result of rising SST’s.  To identify the potential changes, both t-tests and Pearson correlation tests were used to identify change and correlation among the variables defined. The hurricane data used for the analysis was obtained from the National Hurricane Center’s (NHC) HURDAT 2 Atlantic database.  The rationale for using the Atlantic database over the Northeast and North Central Pacific database was due to the availability of more data on hurricanes over a longer period of time.  To create an accurate, average of the data with respect to a yearly perspective, two programs were written to convert the raw data.  The programs took the data and created averages for wind speed (knots), pressure (mb), as well as, assigning the storm an appropriate hurricane category, based on the Saffir-Simpson scale for hurricane categorization.  Based on the Saffir-Simpson scale, tropical storms and depressions were considered tropical cyclones while hurricanes were given their category based on their highest wind speed recorded.  For each record where the wind speed or pressure data had a sentinel value of “-999.00”, it was determined that the relevant data was unavailable and the record was removed from consideration.  The data created from the first program was then used to calculate yearly summaries of the hurricanes recorded.  The SST data used for the correlation coefficients of observable changes was available on the epa.gov website (EPA, 2018).  SST data identifies the annual anomalies ranging from 1880 to 2015—there is no SST data available past 2015.  While earlier data is available, only data after 1970 was used.  It is strongly believed that improvements in technology, for example, satellite imaging, which allows for a clear observation can only improve data reliability.  The 1970-2015 data ranges were based on these two factors, giving the research 45 years’ worth of data from which to conduct the research and data analysis.

Testing Data for Change

 The first series of tests conducted attempted to identify if there existed a change between the frequency of hurricanes and intensity.  This was done by utilizing t-tests for two data sets, hurricane data from 1970-1993 (Set 1) and data from 1994-2015 (Set 2).  The reasoning behind splitting the data was based on the SST data, which showed a slowdown in SST anomaly growth during the early 1990s.  This point of division served the purpose of demonstrating two distinct regions of SST growth to compare hurricane data against each other.  T-tests were used for the mean yearly wind speed average; the mean yearly pressure average; the frequency of category 4 and 5 hurricanes; the frequency of tropical storms and tropical depressions; the frequency of category 1, 2, and 3 hurricanes; and lastly, the frequency of all tropical cyclones (including hurricanes).  After calculating the t-scores, if it was evident that there was a statistical difference in the means, then a Pearson’s correlation coefficient test was done to demonstrate this relationship graphically as well as, identify the numerical strength of the correlation.

Limitations

 As with most research, this effort had several limitations.  First, the SST data could be a consideration as the SST growth was an average for the world’s oceans, meaning that the Atlantic hurricane data is compared to oceans besides its own. It is probable; however, that the increase in SST anomalies for the world’s oceans coincides with Atlantic sea surface temperature data, due to the results found by the experiment.  Another limitation is associated with the hurricane data obtained from the HURDAT 2 database as the data also included hurricanes that did not make landfall.  This means that there were tropical cyclones that were, on average, stronger than other hurricanes that did make landfall as a result of their location. While this could be a limitation on the data used, the experiments were focused on yearly averages of hurricanes, decreasing the likelihood of erroneous data.  Nevertheless, it is recommended that in future studies, landfall hurricanes are separated from hurricanes that failed to make landfall, to ensure more accurate results.

Results

T-test results

 The t-tests performed demonstrated that there was significant change identified in hurricane patterns between Set 1 and Set 2 as observed in Table 1.  Starting with the yearly frequency of category 4 and 5 hurricanes, the t-test showed a statistically significant increase from Set 1 (M = 0.83, SD = 0.82) to Set 2 (M = 1.95, SD = 1.62), t(44) = -2.93, p = .006. The frequency of category 4 and 5 hurricanes increased by 1.12 with a 95% confidence interval between -2.04 and 2.04.  The yearly frequency of tropical storms and tropical depressions demonstrated a statistically significant decrease between the years Set 1 (M = 10.38, SD = 4.45) and Set 2 (M = 8.00, SD = 3.28), t(42) = -2.08, p = .044.  The difference in means was a 22.93% decrease with 95% confidence between the intervals -2.02 and 2.02.  The yearly average wind speeds also demonstrated a substantial statically significant increase between Set 1 (M = 38.24, SD = 4.34) and Set 2 (M = 44.27, SD = 4.49), t(43) = 4.62, p < 0.001. The difference demonstrates a 13.62% increase in wind speed with a 95% confidence interval of -2.02 and 2.02. Average pressure, however, showed no statistically significant change between Set 1 (M = 998.02, SD = 2.60) and Set 2 (M = 997.71, SD = 3.79), t(37) = .3238, p = 0.750. This means that there was no apparent change with a 95% confidence interval of -2.03 and 2.03. The t-test on hurricanes with categories of 1, 2, or 3, failed to show a statistically significant change between Set 1 (M = 4.21, SD = 1.74) and Set 2 (M = 5.14, SD = 2.36), t(39) = 1.51, p = 0.140. While the means of both sets appear different, they fail to demonstrate difference under a 95% confidence interval of -2.02 and 2.02. The last t-test applied to all tropical cyclones also failed to show a statistically significant difference between Set 1 (M = 15.42 SD = 5.04) and Set 2 (M = 15.03, SD = 5.04), t(44) = -.22, p = .828.  This lack of change was determined within a 95% confidence interval of -2.02 and 2.02. In summary, only three variables demonstrated significant change between 1970 and 2015, rejecting H10

The frequency of hurricanes with a category of 4 or 5

The frequency of tropical storms and tropical depressions

The yearly average wind speed of tropical cyclones

Pearson Correlation Results

 With the t-tests presenting what data should be tested for correlation with SST growth, the Pearson correlation coefficients were calculated for the three types of data that demonstrated a statistically significant difference.  A relationship between SST increases and the frequency of category 4 and 5 hurricanes could not be established r(43) = 0.27, p = 0.729. While the first correlation test resulted in H20 being true, the two other tests conducted resulted in H2a being true. The correlation coefficient of the second test showed a moderate inverse relationship between the decreasing frequency of tropical storms and the increase in SST r(43) = -.48 , p < 0.001. The last correlation test performed demonstrated a statistically significant correlation between wind speeds and the increasing sea surface temperatures r(43) = .54,  p < .001.

Discussion

 Based on the predictions made on the effects of sea surface temperature growth on the intensity and frequency of hurricanes, it is evident that climate change’s effect on sea surface temperature affects hurricanes, thus linking the two phenomena. This, however, was not the only conclusion based on the data; the results of the t-tests and correlation show that climate change is not the only factor affecting the yearly averages for hurricanes. The moderate correlations established for wind speeds and tropical storm frequency demonstrate that there exist other factors, affecting the changes.                Climate change may still influence hurricanes beyond just its influence on the rising sea surface temperatures. It has already been demonstrated that UV radiation does influence hurricane patterns, meaning the atmosphere is as important to study as the sea surface temperature.  Furthermore, the reduction of limitations noted prior, would enable better readings on SST growth compared to hurricane patterns.

Conclusion

 In conclusion, climate change has a profound effect on the world’s weather patterns and their effects on natural phenomena such as forest fires or hurricanes.  The results of this experiment validate some of the predictions made, demonstrating a correlation between sea surface temperature anomalies and decreasing tropical storm frequency, as well as, increasing wind speeds. This experiment also demonstrates that there exist other factors that affect hurricanes and that those factors may be linked to climate change as well.  Whether they are linked should be experimented on, to test the current predictions for changes in hurricane patterns.

References

Bedsworth, L., Cayan, D., Franco, G., Fisher, L., & Ziaja, S. (2018). California’s Fourth Climate Change Assessment: Statewide Summary Report(Vol. 4, Rep. No. SUM-CCCA4-2018-013). Retrieved December 5, 2018, from http://www.climateassessment.ca.gov/

Landsea, C., Franlkin, J., & Beven, J. (2017, November 9). HURDAT Re-Analysis Project. Retrieved December 3, 2018, from http://www.aoml.noaa.gov/hrd/hurdat/Data_Storm.html

Elsner, J. B., & Jagger, T. H. (2009). Statistical Link Between United States Tropical Cyclone Activity and the Solar Cycle. In Hurricanes and Climate Change(pp. 61-65). New York City, NY: Springer Science Business Media. doi:10.1007/978-0-387-09410-6

United States of America, Environmental Protection Bureau. (2018, December 17). Climate Change Indicators: Sea Surface Temperature. Retrieved December 3, 2018, from https://www.epa.gov/climate-indicators/climate-change-indicators-sea-surface-temperature

United Nations Framework Convention on Climate Change., Secretariat. (1997). Kyoto protocol(p. 6). Kyoto, Japan: UNFCCC. Retrieved December 4, 2018, from https://unfccc.int/resource/docs/convkp/kpeng.pdf

US Department of Commerce, & National Oceanic and Atmospheric Administration. (2013, May 29). Hurricanes form over tropical oceans, where warm water and air interact to create these storms. Retrieved December 4, 2018, from https://oceanexplorer.noaa.gov/facts/hurricanes.html

Geophysical Fluid Dynamics Laboratory. Global Warming and Hurricanes: An Overview of Current Research Results. (2018, September 20). Retrieved December 4, 2018, from https://www.gfdl.noaa.gov/global-warming-and-hurricanes/

Walsh, J., D. Wuebbles, K. Hayhoe, J. Kossin, K. Kunkel, G. Stephens, P. Thorne, R. Vose, M. Wehner, J. Willis, D. Anderson, S. Doney, R. Feely, P. Hennon, V. Kharin, T. Knutson, F. Landerer, T. Lenton, J. Kennedy, and R. Somerville, 2014: Ch. 2: Our Changing Climate. Climate Change Impacts in the United States: The Third National Climate Assessment, J. M. Melillo, Terese (T.C.) Richmond, and G. W. Yohe, Eds., U.S. Global Change Research Program, 19-67. doi:10.7930/J0KW5CXT.

Knutson, T.R., J.J. Sirutis, G.A. Vecchi, S. Garner, M. Zhao, H. Kim, M. Bender, R.E. Tuleya, I.M. Held, and G. Villarini, 2013: Dynamical Downscaling Projections of Twenty-First-Century Atlantic Hurricane Activity: CMIP3 and CMIP5 Model-Based Scenarios. J. Climate, 26, 6591–6617, https://doi.org/10.1175/JCLI-D-12-00539.1 

Appendix

Table 1:

T-Tests on Tropical Cyclone Patterns between 1970-1993 and 1994-2015

Tropical Cyclone Patterns

Set 1 1970-1993

(n = 24)

Set 2 1994-2015

(n = 22)

t

p

M

SD

M

SD

Yearly Frequency of Category 4 & 5 Hurricanes

.83

.82

1.95

1.62

-2.93

0.00648

Yearly Frequency of Tropical Storms and Depressions

10.38

4.45

8.00

3.28

2.07

0.04703

Yearly Average Wind Speed (knots)

38.24

4.34

44.27

4.49

-4.626

0.00003

Yearly Average Pressure (mb)

998.02

2.60

997.71

3.79

.3238

0.74791

Yearly Frequency of Category 1,2, and 3 Hurricanes

4.21

1.74

5.14

2.36

-1.51

.14014

Yearly Frequency of All Tropical Cyclones

15.417

5.04

15.09

5.10

-0.22

0.82752