Performing Analysis of Meteorological Data

 

 First let us understand meteorology and importance of data analysis on meteorological  data.

"Meteorology is the study of the Earth's atmosphere and the variations in temperature and moisture patterns that produce different weather conditions. Some of the major subjects of study are such phenomena as precipitation (rain and snow), thunderstorms, tornadoes, and hurricanes and typhoons. 

The importance of meteorological events is felt in various ways. For example, a drought results in water shortages, crop damage, low river flow rates, and increased wildfire potential. The critical impact of weather on human activity has led to the development of the uncertain science of weather forecasting. For more information refer https://www.scholastic.com/teachers/articles/teaching-content/meteorology/."

Basically, our aim here is to transform the given raw data into information and then extract our insights and knowledge from it. In this blog, we will perform data analysis on meteorological dataset  available on https://www.kaggle.com/muthuj7/weather-dataset.

So, let's get started.

Hypothesis of the Analysis :

“Has the Apparent temperature and humidity compared monthly across 10 years of the data indicate an increase due to Global warming.”

1)Firstly, we have to import important libraries to be used in this analysis.

 

2)Then, we are using pandas read_csv() function to read our weather dataset. Since the weatherHistory.csv is stored in the same folder(in my pc), so the complete path is not needed. 

 

3)To view first 5 rows, use head() function.

4)Now we want our data to be resampled and in order to do so, we are dropping unwanted data and converting it according to our requirement.

5)Since, we are analyzing data yearly, so we have to convert Formatted Date  into datetime format using pandas method to_datetime(). Furthermore, we will set Formatted Date as the index to the data-set using pandas function set_index().

6)In order to visualize variation or relationships between variables,we have to plot different graphs. We are plotting a graph to display the variation of average apparent temperature and average humidity over 10 years(2006-2016).

 

OBSERVATION: We can conclude from the above plot that humidity remains almost constant throughout 10 years, whereas temperature curve has many ups and downs. Global warming and many other environmental factors can be the cause of this uncertainty in the temperature.

6) Now we are specifically retrieving the data of a particular month, i.e. October by using bar plot.

 

OBSERVATION: It can be clearly seen that the variation of Average Humidity over the year is too small ranging between 0.5–0.8 and the range of Average Apparent Temperature is between 9–14.

7)Here is a plot of the average temperature and humidity in the month of October over the stretch of 10 years.

OBSERVATION: We can observe that in the month of October 2010, the temperature observed is less as compared to the other years. Unlike humidity, temperature has experienced many up and downs over these 10 years.

CONCLUSION: 

We can clearly conclude that there are many rise and fall in temperature over past years(2006-2016), whereas the average humidity remained almost constant throughout these 10 years. The main reason for this drastic variation in temperature is the addition of enormous amounts of greenhouse gases in the atmosphere, thus increasing the greenhouse effect and global warming.