Abstract
Remote sensing through satellites and internet of things (IoT) technology are two widespread techniques to assess inland water quality. However, both these techniques have their limitations. IoT provides point data, which is insufficient to represent entire water body, especially if the water body has complex terrain and hydrology. Through remote sensing, we can sample data of a large area, but data acquisition is constrained by satellite. Revisit time and quality of estimates can be affected by image resolution. Moreover, non-optical properties that might affect water quality cannot be sensed through satellites. To complement this, GIS data from labs can be useful for providing higher resolution and accurate data and can be used as ground truth. Thus, in this chapter, the authors aim to integrate both these data collection techniques followed by estimation and prediction through machine learning models. The accumulated datasets are used to train machine learning (ML) models deployed at a server. The selected ML model is an artificial neural network with train accuracy of 97% and test accuracy of 95%.
TopBackground
Water is an essential resource, despite its pivotal role in the sustenance of life on planet earth, water quality is constantly compromised and degraded by certain human activities. Poor water quality is a problem of great concern worldwide. Particularly in countries like Pakistan, the available water for drinking is impure and polluted by human, agricultural and industrial waste. Poor governance and management in the water monitoring sector further adds to the miseries.
Unfortunately, Pakistan ranks 80th among 122 nations regarding water quality (M.K. Daud., Muhammad Nafees. & Shafqat Ali. (2017) Drinking Water Quality Status and Contamination in Pakistan). The two primary water sources, surface and ground water are heavily contaminated by toxic metals, pesticides, and coliforms. Besides, human factors such as improper disposal of waste, frequent use of agrochemicals, and urbanization have badly affected the water quality in Pakistan, leading to 40% deaths and 50% diseases. Given the lack of water treatment facilities, the water quality problem could get worse in the future if Pakistan’s government continues to overlook this issue. In the current situations, researchers and environmentalists would require real-time and frequent monitoring of quality of a water body, which is often hampered by conventional lab testing. The current lab testing for water quality is a slow, costly and overall inefficient method for a research setup. One of the novel solutions is detecting water quality in real-time through IoT sensors. IoT sensors quickly monitor and report quality parameters saving a great amount of time spent in lab to get the same results. However, IoT provides point data only and could be insufficient to reflect the water quality status of a large water body. Additionally, accessing a complex terrain or harsh areas could be challenging for data collection.
Alternatively, sensing water quality remotely through satellites imagery is a viable technique to gather sufficient data from a large water body. However, remote sensing techniques have their own drawbacks. Firstly, those water parameters that are not optically active cannot be monitored through satellites bands or imagery. Secondly, the data acquisition is bounded by satellite revisit time. Lastly, satellite image resolution, and environmental & climatic effects like cloud covering, back-scattering, top-of the atmosphere reflectance, could significantly influence the results derived from the satellite imagery.
Key Terms in this Chapter
Atmospheric Corrections: It is the process of removing atmospheric effects on the reflectance values of a satellite image.
XML: Extensible markup language or XML is a markup language for encoding documents in a human and machine-readable format.
MLP: Multi-layer perceptron or MLP refers to a feedforward artificial neural network with many hidden layers.
Masking: It is the process of cropping a satellite image based on a shape file.
WQI: A water quality index or WQI is a standard which derives the status of water quality based on given parameters.
Adam’s Optimizer: Adam’s optimizer is a supplement for stochastic gradient decent algorithm which is commonly used for optimization purposes.
GIS: Geographic Information System or GIS is a system for accumulating, managing, and analyzing geo-spatial data.