Tsunami Data Analysis

Dataset Source

The NCEI/WDS Global Historical Tsunami Database, containing tsunami source information and runup data. This dataset includes information on locations where tsunami effects were observed.

Dataset Columns

1. Year: The year the tsunami occurred.
2. Month: The month the tsunami occurred.
3. Tsunami Event Validity: Validity of the actual tsunami occurrence, rated from -1 (erroneous) to 4 (definite).
4. Tsunami Cause Code: The source of the tsunami, from 0 (unknown) to 11 (astronomical tide).
5. Deposits: Criteria commonly used to identify tsunami deposits.
6. Country: Country name where the tsunami occurred.
7. Location Name: The area along the coastline.
8. Number of Runups: Total number of locations along the coastline where the tsunami wave reached a maximum height above the normal sea level.
9. Earthquake Magnitude: Size or strength of the earthquake (Richter Scale).
10. Latitude and Longitude: The location where the tsunami originated.
11. Maximum Water Height: Maximum elevation above the normal sea level that a tsunami wave reached at a specific location.
12. Tsunami Intensity: Potential impact of a tsunami event, calculated using runups height.
13. Total Damage Description: Ranges from 0 (none) to 4 (extreme).
14. After 1900: A binary column indicating if the year is before (0) or after (1) 1900.

Data Cleaning

The original dataset had 1433 rows and 47 columns, with many NaN values. After cleaning, the dataset was reduced to 280 rows and 15 columns.

Statistical Analysis

1. Multiple Linear Regression: Examined the relationship between all variables (excluding Year, Month, Latitude, Longitude, and After 1900) and Tsunami Intensity. The model was significant with an R² of 18.65%, and significant variables were Deposits, Earthquake Magnitude, Maximum Water Height, and Total Damage Description.
2. Factor Analysis: Reduced data into factors:

• Factor 1: Effects of Tsunami (Deposits, Number of Runups)
• Factor 2: Timeline (Year, After 1900)
• Factor 3: Tsunami Cause Code
• Factor 4: Month, Tsunami Event Validity
• Factor 5: Location (Latitude and Longitude) The dataset was saved as Tsunami_Factors with 280 rows and 20 columns.

3. Multiple Linear Regression with Factors: Significant model with an R² of 55.42%. All factors had a significant impact on Tsunami Intensity except Maximum Water Height and Earthquake Magnitude, with Factor 1 having the highest impact.
4. Independent Sample t-test: Compared average Tsunami Intensity before and after 1900. The result showed no significant difference, implying no impact of climate change on tsunami intensity.

How to open ctk file

Download SAS Studio to open the ctk files and view the analysis done on the dataset with selected options or code.