Groundwater sustainability is critical to the future of agriculture and food security. The challenges are not only technical but have important social, economic, institutional and policy implications. The objective of this research is to predict groundwater levels in rural wells, allowing farmers to use their groundwater more sustainably. Data visualizations and machine learning algorithms are used to examine data collected over a five-year period from rural rock water basins in the northwestern part of India. Preliminary examination shows that the weekly collected time variable proved to be the single most valuable predictor of groundwater level, as it included implied seasonal changes in weather patterns and pumping patterns. However, due to limited rainfall outside of the monsoon season, it proved a less potent variable than previously expected.
Stocker, Matthew; Iyer, Lakshmi; Maheshwari, Basant; and Sharda, Ramesh, "Predicting Groundwater Fluctuations in Hard Rock Watersheds – An Application of Data Visualizations and Machine Learning Algorithms" (2020). Proceedings of the 2020 Pre-ICIS SIGDSA Symposium. 6.