The water quality index (WQI) is an essential indicator to manage water usage properly. This study aimed at applying a machine learning-based approach integrating attribute-realization (AR) and support vector machine (SVM) algorithm to classify the Chao Phraya River's water quality. The historical monitoring dataset during 2008-2019 including biological oxygen demand (BOD), conductivity (Cond), dissolved oxygen (DO), faecal coliform bacteria (FCB), total coliform bacteria (TCB), ammonia (NH 3 -N), nitrate (NO 3 -N), salinity (Sal), suspended solids (SS), total nitrogen (TN), total dissolved solids (TDS), and turbidity (Turb), were processed via four studied steps: data pre-processing by means substituting method, contributing parameter evaluation by recognition pattern study, examination of the mathematic functions for quality classification, and validation of obtained approach. The results showed that NH 3 -N, TCB, FCB, BOD, DO, and Sal were the main attributes contributing orderly to water quality classification with confidence values of 0.80, 0.79, 0.78, 0.76, 0.69, and 0.64, respectively. Linear regression was the most suitable function to river water data classification than Sigmoid, Radial basis and Polynomial. The different number of attributes and mathematic functions promoted the different classification performance and accuracy. The validation confirmed that AR-SVM was a potent approach application to classify river water's quality with 0.86-0.95 accuracy when applied three to six attributes.

Inoculum is a crucial factor influencing the success of an anaerobic digester start-up. This study evaluated the efficacy of aerobic waste activated sludge (WAS) to start-up anaerobic hybrid reactors (AHR) treating cassava wastewater. The start-up performance and stability of five 6.2 L AHRs inoculated differently by WAS (AHR1) and WAS co-inoculated with pig manure (AHR2), cow dung (AHR3), chicken manure (AHR4), and anaerobic pond sludge (AHR5) were investigated. The results depicted the potential of WAS as starter seed for AHR's start-up. This sludge contained methanogenic populations, mainly Methanomicrobials, Methanosarcinales, and Thermoplasmatales, for 1.5 × 10 7 MPN/g VS compared to 4.0x10 3 -6.0x10 7 MPN/g VS in manures and anaerobic sludge. Co-inoculation of WAS with cow dung (AHR3) and anaerobic sludge (AHR5) promoted a rapid start-up, achieved 4.0 kg COD/m 3 .d within 1.5-1.7 months, while about 2.5-3.3 months needed for the others. A satisfying performance, stability, microbial development and activity, depicted as high COD removal (COD rem ) efficiency and methane productivity, were achieved in all digesters. About 0.32 and 0.21 m 3 /kg COD rem for biogas and methane yielded in AHR1, whereas 0.41-0.49 m 3 -biogas/kg COD rem and 0.25-0.32 m 3 -CH 4 /kg COD rem obtained in other reactors. This finding confirmed that inoculum's quality is critical for productive gas productivity and system stability maintenance.

Turbidity is a standard water quality parameter that indicates its optical property in scattering light along the column containing suspended particles. The satellite imagery information of Sentinel-2 and the Chao Phraya River turbidity data from December 2016 to February 2021 was applied to develop a mathematical equation for turbidity determination. This practical and straightforward approach eliminates some constraints of traditional laboratory analysis, which is labour-intensive and time-consuming in monitoring the entire river. Four studied steps were implemented: data pre-processing, correlating analysis of numerical turbidity and satellite image reflectance, developing the mathematic equations for turbidity estimation, and its validation of use. Four different bands (B2, B3, B4, and B8) and three selection methods were investigated; single-band, combination band, and ratio band. The obtained results depicted that the reflectance of B4 in the single-band process promoted the highest correlation with turbidity compared to the others. The reflectance in visible wavelengths increased when the turbidity of river water increased, particularly B4. The mathematical power equation was a more suitable function for evaluating turbidity than linear regression, quadratic, and exponential functions. A similar concentration was obtained for measured and estimated turbidity in the validation. This finding demonstrated the potential application of remotely sensed data to estimate river water turbidity with high capability and accuracy that adequately supports spatial data continuity acquisition.