Testing linear regression to predict the influence of macro-scale parameters on micr-scale water quality parameters

Master thesis (2023)

Authors

A.W. Dommerholt Civil Engineering & Geosciences

Contributors

M.W. Ertsen Water Resources - CEG (supervisor 1)
R.S. Houser Water Resources - CEG (supervisor 2)
S. Pande Water Resources - CEG (supervisor 2)
Faculty
Civil Engineering & Geosciences
Copyright: © 2023 Naomi Dommerholt
More Info
expand_more

Published Date

05-10-2023

Language

English

Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Faculty

Civil Engineering & Geosciences

Abstract

This study investigates the connection between macro-scale parameters (MSP) and micro-scale water quality parameters (MSWQP) in the Brantas River, employing a multivariate linear regression (MLR) modelling approach. The analysis reveals several critical insights into the complex dynamics of water quality in this river system.

Key findings indicate that high-quality input data, both in terms of quantity and measurement frequency, play a key role in the effectiveness of predictive models. Seasonality is a useful predictor and is recommended to be supplemented with rainfall data to better capture its influence on runoff and water quality. The study introduces the concept of basin accumulation and the implementation of buffer areas, demonstrating that these enhancements lead to improved model performance.

In conclusion, it can be said that relying just on macro-scale parameters is insufficient to generate an effective linear regression model. However, with the right optimizations and useful input data, it can be an insightful and valuable tool for water quality prediction.