Rainfall Data (Unpivoted) [Need Visualization] {Script for unpivoting done}

[Kshithij Iyer]

DataMeet Pune Rainfall Data (Unpivoted)

This is a script which I developed for DataMeet Pune which takes an .xls file which has rainfall data of India and generates state-wise .csv files in an unpivoted format which can be useful for further analysis

.

What does the script do?

Look at the unpivoting example.xlsx file and this link you'll get a hang of what the script does. :)

About the data?

Source of Data: http://www.indiawaterportal.org/articles/district-wise-monthly-rainfall-data-list-raingauge-stations-india-meteorological-department

Issues with the Raw data: Chandigarh data is given under Chhattisgarh sheet. A sepreated sheet had to be manually created so that the script can run smoothy on the .xls file.

List of R packages needed by the code:

1. xlsx
2. tidyr

Manual inputs needed to the script.

1. You'll have to choose the .xlsx file which has to be processed. However the code does have manual logic as well which can be uncommented and you can create .CSV files of specific sheets instead of the all the sheets in the .xls file.
2. The .csv file would be store in the ./Processed-Data folder and You'll have to manually add "Sr.No", to the start of the .csv file.

What can be done with this data?

1. A choropleth map of the country with a timeline, the colours changing to show amount of rainfall.
2. A predective model for drought detection can be created.

[Sandeep Kumar]

I think that is pretty easy in pandas also i have analysed daily rainfall for jaipur for nearly five years here https://github.com/sandeepgadhwal/Rainfall-and-Stormwater-Analysis/blob/master/Rainfall_Analysis.ipynb