<![CDATA[Water pollution has become a complex environmental challenge in rapidly urbanizing Southeast Asia. Indonesia has deployed over 347 online river water quality monitoring stations (ORWQM) in several watersheds to provide real-time data for pollution control and informed ecological decision-making. However, problems have arisen with some of the longstanding stations, including equipment degradation, data loss, sensor malfunctions, and inconsistencies, raising concerns about the reliability of these datasets. This study aims to address these issues by using a hybrid Total Data Quality Management (TDQM), Online Data Quality Evaluation Model (ODQEM) framework to evaluate and improve the data quality at four stations (Manggarai, Kelapa Dua, Pasar Baru, and Empang Dam) representing the upstream and downstream areas of the Ciliwung and Cisadane Rivers. The workflow consists of four steps: (i) defining the relevant data quality dimensions; (ii) measuring data quality using completeness index, range validity, accuracy, uniqueness, and stability index; (iii) analyzing cross-parameter and cross-station patterns to identify potential failure modes and governance gaps; and (iv) translating findings into targeted calibration, cleaning, and maintenance actions. The results show strong structural integrity; completeness indicators reach 100%, and uniqueness exceeds 99.9%, indicating robust acquisition and temporal consistency. However, functional reliability varies widely across parameters and station locations. Completeness indicators for non-zero values show systematic zero values for the DO probe (1.02%) and nitrate (0.19%) at Empang Dam Station; TDS (9.57%) at Pasar Baru Station; ammonia (60.02%) at Kelapa Dua Station. Overall stability is high for physical probes (temperature, pH), but low for chemical/ion probes at some stations (salinity 1.47%–6.62%; TDS 5.49%–10.92%; ammonia 30.20%–71.40%). Ion sensors also show higher risks, including low validity or accuracy for nitrate at Kelapa Dua Station (61.67%). These findings indicate that the dataset appears complete and valid, but still contains substantial functional uncertainty due to several zero-value parameters and unstable sensor behavior. Because the evaluation is limited to four stations, these findings should be interpreted as an in-depth case study rather than a complete representation of the ORWQM network in Indonesia. The proposed TDQM–ODQEM Data Governance Model offers a replicable method for improving online and real-time environmental monitoring, helping bridge the gap between technological performance and policy design in sustainable river management.]]>
