Climate Change and Adaptation Data Analysis: Queensland Coast Report

This report presents a data analysis of climate change impacts on the Queensland coastline. The study examines the relationship between sea level rise, temperature changes, rainfall patterns, and the frequency of tropical cyclones. Utilizing statistical methods such as linear and multiple regression, the analysis reveals a negative correlation between the number of tropical cyclones and the passage of time, suggesting a potential decrease in their impact. The study also explores the influence of temperature and rainfall on sea level rise, indicating that temperature has a more significant impact than rainfall. The findings highlight the importance of understanding these climate variables to develop effective adaptation strategies for coastal communities in Queensland. The report concludes with a summary of the key findings and references relevant data sources from the Australian Government Bureau of Meteorology.

Running head: CLIMATE CHANGE AND ADAPTATION DATA ANALYSIS 1
Climate Change and Adaptation Data Analysis
Name
Institutional Affiliation

CLIMATE CHANGE AND ADAPTATION DATA ANALYSIS 2
Data Analysis
There are various factors that leads to rise in sea level along the Queensland coastline.
This statistical analysis attempts to analyze the effects of global temperature change, changing
intensity of storm tides and cyclones and the changing rainfall pattern as factors that contribute
to the rise in sea level of the Queensland coastline. Temperatures and rainfall amounts are some
of the factors that affect changes in sea level that subsequently lead to flooding along the
Queensland coastline.
Rise in temperatures causes snow melt and weakening of water molecules causing
increase in volumes of the ocean water. As well, higher annual rainfall amounts also add to the
ocean water volume thus causing increased sea level. Tropical cyclones and storms may also
affect the structure of the coastline as they may erosion. To prove this the temperature and
rainfall data from the Queensland Gold coast sea way weather station were obtained and
analyzed to see how these factors are related to the rise in sea levels.
1. Analysis of number cyclones from 1995 – 2017 by simple linear regression
Summary descriptive statistics
Mean 9.863636364
Standard Error 0.600488454
Median 10
Mode 10
Standard Deviation 2.816540508
Sample Variance 7.932900433
Kurtosis 0.825747265

CLIMATE CHANGE AND ADAPTATION DATA ANALYSIS 3
Skewness -0.367581064
Range 12
Minimum 3
Maximum 15
Sum 217
Count 22
For the 22 years, it is evident that the mean number of tropical cyclones is about 10 with
a standard deviation of 3. The least experienced number of cyclones is 3 and the maximum is 15.
The skewness of -0.37 indicates that the data is fairly symmetrical
Summary output of Regression analysis
Regression Statistics
Correlation coefficient -0.56368693
R Square 0.31774296
Adjusted R Square 0.28363011
Standard Error 2.3838815
Observations 22
ANOVA
df SS MS F Significance

CLIMATE CHANGE AND ADAPTATION DATA ANALYSIS 4
F
Regression 1 52.933088 52.933088 9.314465 0.0062933
Residual 20 113.65782 5.6828910
Total 21 166.59090
Coefficien
ts
Standard
Error t Stat P-value
Lower
99.0%
Upper
99.0%
Intercept 500.19762 160.662759 3.1133389 0.005477 43.057501 957.33775
X Var 1 -0.244494 0.08011067 -3.0519608 0.006293 -0.4724367 -0.0165526
There exists a medium negative relation of tropical cyclones across the years as shown by
the correlation coefficient of -0.56. There is about 28.4% variation explained by the change in
years that affect the number of tropical cyclones. In hypothesis testing, we test if the coefficient
of X variable 1 is zero. The null hypothesis is H0 = 0, and alternative hypothesis is H0 ≠ 0.
From the ANOVA test results, at 0.01 the significance level, the p value is considered
less than the alpha level hence the null hypothesis is rejected. We can conclude that the
coefficient (slope) of X variable is therefore significantly different from zero and thus there
exists a relationship between these two variables.
About 31.8% of the data can be explained by the regression line with equation;

CLIMATE CHANGE AND ADAPTATION DATA ANALYSIS 5
Y= 500.20 – 0.24X
where y is the projected number of cyclones and X is the year. This means that as the years
increase, there is a high probability of the cyclones reducing. This indicates that tropical cyclones
may not have adverse erosion and displacement effects on the Queensland coastline since the
numbers reduce with time.
With 99% confidence, we can say that the value of H0, mean slope for the two variables
lies between -0.472 and -0.017. There is a 99% chance that the true rate of decrease in the
number of tropical cyclones with increase in years lies within these two values. This implies that
if some other samples were to be analyzed, there would be a 99% chance that the slope of their
linear regression line would be between the variables would be between these two intervals.

CLIMATE CHANGE AND ADAPTATION DATA ANALYSIS 6
2. Analysis of rise in sea level verses annual temperatures and rainfall amount by multiple
linear regression.
Summary output of descriptive characteristics
Temperature(0c) Rainfall (mm) Rise in sea level (m)
Mean 25.463636 Mean
1273.48
2 Mean 0.81871
Standard Error 0.1869724 Standard Error
63.9907
6 Standard Error 0.012947
Median 25.75 Median 1266.9 Median 0.822583
Mode 25.8 Mode #N/A Mode #N/A
Standard
Deviation 0.8769783
Standard
Deviation
300.143
3
Standard
Deviation 0.060727
Sample Variance 0.7690909 Sample Variance
90085.9
7 Sample Variance 0.003688
Kurtosis -0.73976 Kurtosis
0.43593
2 Kurtosis 0.248948
Skewness -0.699656 Skewness 0.49151 Skewness -0.78993
Range 2.7 Range 1189.4 Range 0.228667
Minimum 23.9 Minimum 832.4 Minimum 0.6765
Maximum 26.6 Maximum 2021.8 Maximum 0.905167
Sum 560.2 Sum 28016.6 Sum 18.01163
Count 22 Count 22 Count 22

CLIMATE CHANGE AND ADAPTATION DATA ANALYSIS 7
For the past 2 decades, temperature values range between 23.90c and 26.60c having a
mean of 25.50c with a standard deviation of 0.90c. Such temperatures are so high and can lead to
snow melt and also rise in oceanic heat content weakening the water molecules.
The mean annual rainfall is about 1273mm i.e. about 3.5mm daily rainfall with a
standard deviation of 300mm. the range is between 832mm and 2022mm, the highest being
during 2010 El Nino event. This is an indicator that El Nino and la Nino events have a great
contribution in changes in rainfall patterns.
The estimated annual average increase in sea level is approximately 0.82m with a
standard deviation of 0.06m. By empirical rule, 99% range of rise in sea level is estimated to be
between 0.64m and 1m. The skewness of the all the three variables is (-0.7, 0.5 and -0.8)
indicating that these data are moderate distributed and can be said to be viably good for analysis.
Summary output of Regression Analysis
Regression Statistics
Multiple R
0.67923
98
R Square
0.46136
6706
Adjusted R Square
0.40466
8465
Standard Error 0.04685

CLIMATE CHANGE AND ADAPTATION DATA ANALYSIS 8
5676
Observations 22
ANOVA
df SS MS F
Significance
F
Regression 2 0.03572984 0.01786 8.137231 0.00280082
Residual 19 0.041713633 0.0022
Total 21 0.077443474
Coefficients
Standard
Error t Stat P-value Lower 99.0%
Upper
99.0%
Intercept -0.41891718 0.30958147 -1.3532 0.191883 -1.30460952 0.466775
X Variable 1 0.045795494 0.011782679 3.88668 0.000993 0.01208602 0.079505
X Variable 2 5.61512E-05 3.44274E-05 1.631 0.119355 -4.2343E-05 0.000155
The multiple regression correlation coefficients = 0.68 showing that these variables are
strongly related. About 40.47% variation explained by the change in temperatures and rainfall
amounts that affect the sea level changes.
The coefficient of determination, R squared is 0.46 showing that about 46% of these
variables can be explained by a linear regression model. The adjusted R squared reduces when

CLIMATE CHANGE AND ADAPTATION DATA ANALYSIS 9
rainfall variable is added indicating that rainfall has less effect in sea level rise. The major
contributions of rainfall that can be said to cause rise in sea levels are the El Nino and La Nino
events.
The residual sum of squares is hence close to zero thereby indicating that there is a little
amount of variance from the data that is not explained by the model. The majority of the data is
close to the regression best fit line with the equation;
Y = 0.046X1 + 0.00006X2 – 0.419
where y is the predicted rise in sea level in meters, X1 is the annual temperature in degrees
centigrade and X2 is the rainfall amount in mm.
This equation shows that;
1. When the rainfall variable is held constant, the projected increase in the sea level rise for
every unit increase in temperature is 0.046.
2. When the temperature variable is held constant, the projected increase in sea level rise for
every unit increase in rainfall amount is 0.00006 which is almost insignificant.
3. When all the two factors that can affect sea level rise are zero, there is a decrease in sea level
by 0.419.
In this scenario, we conduct two hypotheses testing i.e. for X1 and X2. We test if the
coefficients of X variables are zero at 0.01 alpha levels. The null hypothesis is H0 and H1= 0, and
alternative hypothesis is H0 and H1 ≠ 0.
The ANOVA test results indicate a significance of 0.0028 of the regression line. This
value is less than the alpha level indicating that there is a general relationship between these

CLIMATE CHANGE AND ADAPTATION DATA ANALYSIS 10
variables. Individual variable hypothesis results indicate that the rainfall variable is zero as the p
value is greater than the 0.01.
A 99% confidence interval test also reveals that the coefficient of the independent
variable, temperature, lies between 0.012 and 0.080. There is a 99% chance that the true rate of
increase in rise of sea level with the changing temperatures is within these two values.
Similarly, for the independent variable rainfall, we can be 99% confident that the
coefficient of this variable lies between -0.00004 and 0.00016. This shows that any kind of data
sample within the Queensland coastline would indicate the rainfall is almost insignificant in
determining the changing sea levels as it coefficient is very small even at the maximum level of
significance.
Conclusion
In conclusion, there are several contributions of rise in sea level and damage to the
Queensland coastline such as temperature changes and severe rainfall events. However, tropical
cyclones may not be of great influence to the coastline structure as the number and severance is
decreasing with yearly changes.

CLIMATE CHANGE AND ADAPTATION DATA ANALYSIS 11
References
Australian Government Bureau of Meteorology. (2019). Monthly Rainfall - 040764. Retrieved
from http://www.bom.gov.au/jsp/ncc/cdio/weatherData/av?
p_nccObsCode=139&p_display_type=dataFile&p_startYear=&p_c=&p_stn_num=04076
4
Australian Government Bureau of Meteorology. (2019). Mean Maximum Temperature - 040764.
Retrieved from http://www.bom.gov.au/jsp/ncc/cdio/weatherData/av?
p_nccObsCode=36&p_display_type=dataFile&p_startYear=&p_c=&p_stn_num=040764
Australian Government Bureau of Meteorology. (2019). Monthly sea levels for Gold Coast -
1987 to 2017. Retrieved from
http://www.bom.gov.au/ntc/IDO70000/IDO70000_60050_SLD.shtml
Australian Government Bureau of Meteorology. (2019). Tropical Cyclone Trends. Retrieved
from http://www.bom.gov.au/cyclone/climatology/trends.shtml