Group04 Report
Water For Life: India's Rainfall & Crop Analysis Through Visualizations
|
|
|
|
|
|
Contents
Introduction
This study focuses on exploratory analysis of rainfall pattern, crop productivity and effect of rainfall pattern changes on crop productivity across different meteorological subdivisions of India. Considering a wide range of weather conditions across a vast geographic scale and varied topography, it won’t be wise to generalise climate changes and its effect on crop productivity in varied regions of India, that is why we decided to explore and visualise rainfall pattern changes and its effect on crop productivity for every meteorological subdivision.
Our research focuses on 34 out of 36 meteorological subdivisions rainfall and crop productivity. Two subdivision excluded are Lakshadweep and Andaman & Nicobar subdivisions. Rainfall data used for research is at meteorological subdivisions level however crop data used is at administrative district level. We have maintained same granularity level for visualisation of separate data. For analysing effect of rainfall over crop productivity, district data is aggregated to subdivision level based on mapping of subdivisions and districts.
Crop growing season in India is classified into two main seasons – (i) Kharif and (ii) Rabi based on monsoon. The Kharif cropping season if from July- Oct during south– west monsoon and Rabi cropping season is from October- March (Winter). Crop grown between March- June are Summer Crops. Apart from these seasonal crops there are few crops which are grown throughout year and are classified as Whole year crops. So, we have considered total Four seasons as Kharif, Rabi, Summer and Whole Year for analysis.
We have used multiple visualisation technics such as Time series, Heat Map, Tree Map, Parallel Coordinate Plot, Geo Facet and Bar charts for easy visualisation of varying rainfall pattern and changing crop productivity for 15 years across different subdivisions of India. R is used for data visualisation and application as it offers satisfactory set of inbuilt functions and libraries for both data mining and visualisation.
|
Objective And Motivation
Climate plays a significant role in economic development of India. Because large population of India depends on climate sensitive sectors like agriculture and forestry for livelihood. Climate change could lower the farmer’s income by up to 25% (Economic Survey 2018: http://mofapp.nic.in:8080/economicsurvey/pdf/082-101_Chapter_06_ENGLISH_Vol_01_2017-18.pdf).This is because agriculture in India is vulnerable to the vagaries of whether as close to 52% farm land is still unirrigated and depends on rainfall. This project is honest endeavor in gaining deeper knowledge into the impact of increasingly changing rainfall patterns, so that we can be prepared to mitigate the risk of these uncontrollable factors and seek remedies that would help sustain such drastic natural phenomenon.
Considering crops cultivation period and water requirement for crop during different stages of its lifecycle, it is important to analyse effect of monthly rainfall on crop productivity during cultivation period rather than simply considering yearly/seasonal average rainfall. Our objective is to provide single view to analyse monthly rainfall pattern changes, crop productivity changes and correlation between every month’s rainfall and crop productivity.
|
About The Data Source
|
Critique of the Existing Visualizations
Dashboard Design
|
Key Insights
Conclusion/Future Work
Given time constraints and the nature of data we gathered, this application is only limited to show the correlation between crop productivity and rainfall pattern changes. We cannot conclude rainfall pattern change is the causation for crop productivity change in Indian agricultural sector, as there are several other factors impacting the cultivation and harvesting of various crops in different regions of India such as temperature, wind, soil as well as capital and government support.
|
Acknowledgement
We would like to extend our gratitude towards Dr Kam Tin Seong (Singapore Management University) for his guidance on analytical techniques and R packages that may be used and feedback on visualisation techniques. Without his encouragement and technical assistance, this project would not be as it is today.
|
R Packages Used
We have used the following R packages to come up with our visualizations: |
