We were inspired by the challenge of finding innovative solutions to address water scarcity using artificial intelligence (AI). We wanted to create a project that could help people understand the current and future state of water availability in their region, and provide them with useful resources and tips to conserve water and reduce their water footprint. We also wanted to raise awareness about the causes and effects of water scarcity, and encourage people to take action and join the global movement for water justice.
Our project predicts drought and water scarcity in different regions of India, using a seasonal autoregressive integrated moving average (SARIMA) model. The model takes into account the historical data of rainfall, temperature, and vegetation index, and forecasts the future values of these variables. Based on the forecast, the model assigns a drought severity index and a water scarcity level to each region. Our project also provides an interactive map, a dedicated resources page, a search function for checking water status, social media integration, and engaging graphics. These features aim to help users learn more about water scarcity, find ways to save water, and share their stories and opinions with others.
We built our project using a combination of technologies, such as satellite imaging, government datasets, machine learning, web development, and data visualization. We used satellite images from the NASA Earth Observations to measure the normalized difference vegetation index (NDVI), which is an indicator of plant health and water content. We also used datasets from the World Bank and the World Resources Institute to obtain information on water resources, water use, water stress, and water risk. We applied machine learning techniques, such as regression and classification, to predict the areas prone to drought and the level of water scarcity in different cities. We developed a web application using Streamlit, a framework for creating data-driven apps, to display our results and features. We used various libraries and tools, such as Pandas, NumPy, Matplotlib, Plotly, and Folium, to process, analyze, and visualize our data.
Some of the challenges we faced during the project were:
Finding reliable and updated data sources for water scarcity and drought. Processing and aligning the satellite images with the geographic coordinates of the regions of interest. Choosing and tuning the appropriate machine learning models for our prediction tasks. Designing and testing the user interface and functionality of our web application. Managing the time and workload of our team members and ensuring effective communication and collaboration. Accomplishments that we’re proud of We are proud of the following accomplishments:
We successfully implemented a SARIMA model to forecast drought and water scarcity in different regions of India, with a high accuracy and reliability. We created a user-friendly and informative web application that showcases our results and features, and provides a platform for users to interact and learn. We used satellite images and data visualization to create engaging and meaningful graphics that illustrate the state of water availability and the impact of water scarcity. We learned a lot from this project, both in terms of technical skills and domain knowledge. What we learned We learned a lot from this project, both in terms of technical skills and domain knowledge. We gained experience in working with satellite images, machine learning models, and web development tools. We also learned more about the causes, effects, and solutions of water scarcity, and the importance of water for life on Earth.
What’s next for Drought Prediction using SARIMA We plan to extend our project in the following ways:
We want to include more data sources and variables, such as soil moisture, groundwater, and evapotranspiration, to improve our model and predictions. We want to expand our scope and scale, and apply our model to other regions and countries that are facing water scarcity issues. We want to add more features and functionalities to our web application, such as a personalized dashboard, a feedback system, and a gamification element. We want to reach out to more users and stakeholders, and collaborate with them to create a positive impact and a lasting change.