Time Series Analysis and Prediction of Water Quality through Remote Sensing Geoinformation System and Internet of Things

The periodic time series analysis of different aspects of urban areas is essential owing to the increase in population, stressed resources, and lack of technology-based solutions for effective resource management. In this regard, the temporal analysis of water quality holds paramount importance as i

Project Title

Time Series Analysis and Prediction of Water Quality through Remote Sensing Geoinformation System and Internet of Things

Project Area of Specialization

Artificial Intelligence

Project Summary

The periodic time series analysis of different aspects of urban areas is essential owing to the increase in population, stressed resources, and lack of technology-based solutions for effective resource management. In this regard, the temporal analysis of water quality holds paramount importance as it is a key ingredient to sustain life on earth. Water quality depends on many factors like precipitation, climate, soil type, vegetation, geology, flow conditions, groundwater, PH level, etc. For this purpose, the remote sensing data from satellites (preferably Sentinel and Landsat), Geographic Information System (GIS) data, and IoT data will be collected to perform water quality trend analysis. The data will be used to generate water quality parameters. A water quality index will be generated against each set of water quality parameters to determine the quality of the water. The machine learning classification and regression techniques and algorithms will be used to classify, analyze, and predict the water quality. The subject area for this study will preferably be the Rawal dam. The analysis will be performed on the data of the last ten years and predictive analytic will be performed to project the water quality for the near future. This analysis would help us to identify various factors (such as increased air pollution, weather conditions) that are contributing to polluting the surface water and how remedial measures can be put in place to mitigate the decreasing water quality. The time series analysis & visualization and predictive analytic service will be developed on a web portal.

Project Objectives

The main objectives of the project include:

To collect data on factors affecting the water quality of the last decade using Remote Sensing, Geographic Information System, and the IoT equipment.

To preprocess the data and generate water quality parameters from it.

To generate water quality index using the data of water quality parameters.

To classify water quality using machine learning algorithms.

To predict the water quality by doing time-series analysis.

To identify the key sources causing a reduction in water quality.

To develop a web portal to display time series analysis and results.

Project Implementation Method

The project will basically be implemented using machine learning classification and regression algorithms and techniques using python libraries. The data will be collected using Landsat and Sentinel satellites from Earthdata from year 2011-2020. The data will be preprocessed using the shapefile of our area, i.e. Rawal Lake. The tools like SNAP and libraries like Xarray, NetCDF will be used. After preprocessing, the bands' data will be used to calculate water quality parameters. After data collection from satellites, the data will be collected using IoT equipment from in-situ samples. The datasets from Rawal lake filtration plants and PCRWR will also be used. The data will be cleaned and a water quality index will be generated against each set of water quality parameters. After the preparation of the dataset, a machine learning algorithm, such as a support vector machine, long-short term memory etc., will used to train and test the data. After successful training and testing, the model will be applied to predict the value of the water quality index and individual water quality parameters for the new future. A web portal using HTML, Bootstrap, node.js, and javascript will be developed to display and visualize prediction on charts and heat maps.

Benefits of the Project

The research conducted on water quality analysis and prediction has a lot of gaps. In Pakistan, no studies and research have been conducted to predictive analysis of water quality in coming years and There has been no time series analysis. For water quality analytical purposes, only recent data considered. In such cases, the study fails to provide long-term results and a big picture for the future. Other than this, there has a very less use of deep learning algorithms in the analytic study of water quality even though they show promising results. All the studies either involve RS, GIS, or IoT.

Keeping in view our project will cover all these gaps. The collecting data through all these techniques will help in developing a reliable and good dataset. The development of a web portal is also required to display time series and predictive analysis which will be done in our project. It will help in the identification of key water pollutants which will be useful in taking remedial measures against increasing water pollution. This project will benefit people by making them understand the impact of using contaminants like plastic on water quality. It will also help authorities in creating awareness among people to keep the water and environment clean. This study is going to help the government in planning a solution to increase water quality by indicating of the severity of water pollution This research is also aligned with United Nations Sustainable Goal 6; "Ensure availability and sustainable management of water and sanitation for all". Lastly, the data collected will be useful for future research and projects.

Technical Details of Final Deliverable

The final deliverable of the project will be a web portal that will display the time series analysis and predictions made from the data using machine learning regression and classification algorithms. The web portal front-end will be designed using HTML and bootstrap and the back-end will be designed using Node.js and javascript. The data analysis and predictions will be displayed using line charts, heatmaps, correlation matrices etc. The web portal will display the data of water quality parameters of the past decade. The web portal will also display the predictions of water quality parameters for the near future. It will indicate how water quality will be affected in the future and what parameters will be changing the most.

Final Deliverable of the Project

Software System

Core Industry

IT

Other Industries

Others

Core Technology

Artificial Intelligence(AI)

Other Technologies

Others, Big Data

Sustainable Development Goals

Good Health and Well-Being for People, Clean Water and Sanitation, Climate Action

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com