The Chemours plant in Dordrecht discharged PFOA to the River Lek from the 1970s to 2012, reportedly up to about 45,000 kg per year. PFOA is part of the PFAS family, the so called “forever chemicals”. Since they are very persistent, they are harmful to the environment and to people. PFAS-contaminated surface water can infiltrate into polder groundwater. This affected the Lekkerkerk riverbank-filtration area and led to contamination of drinking water. An earlier study by Oasen and Rijkswaterstaat used Oasen’s groundwater model with an average travel time to reconstruct PFOA in the River Lek. Because this approach was simplified and did not include an explicit retardation term, follow-up research was needed.

This study aims to determine historical PFOA levels in the River Lek and assess whether applying retardation improves reconstructed concentrations. A groundwater model of the Lekkerkerk extraction area was developed to estimate travel-time ranges and probability distributions for 18 wells at the Lekkerkerk–-Tiendweg site. The site-specific model includes a detailed geohydrological description and realistically represents near-surface flow and mixing affecting water reaching the wells. Using travel-time distributions provides a more accurate representation by capturing both fast and slow flow paths and the arrival of low concentrations. These distributions were applied in an inverse transfer function to optimise reconstructed river concentrations against well observations.
For chloride, both river and polder data were available along with extensive well observations, while PFOA data existed only for 2016 and 2025. The method was first tested with chloride to assess the effect of data density on accuracy, then applied to PFOA with a retardation factor. It was also tested on PFNA and PFOS to evaluate chain-length effects, and results were compared with the earlier Oasen reconstruction. The reconstructed PFOA history peaks just after 1990 —about a decade earlier than the mean-travel-time approach. A small fitted retardation (R = 1.033) slightly improves the fit, while using a literature value (R = 1.24) delays and worsens it. Main limitations are data sparsity and the smoothing parameter $lambda$, which affect peak timing and magnitude. Using full TTDs rather than a mean is essential for realistic results. Additional monitoring and explicit sorption modelling would further reduce uncertainty and improve concentration and retardation estimates.

Results show that PFOA peaks in the River Lek have passed, but the contaminant persists in the polder, with breakthrough in outer wells and concentrations around 65 ng/L well above the proposed SCHEER limit of 4.4 ng/L. Beyond the studied area, PFAS may spread downstream and return inland via sea spray, affecting dune groundwater and other drinking water sources.
This study demonstrates that TTD-based convolution methods work well for tracers with abundant data and provide useful insight for PFAS when data are sufficient. Reliable PFAS reconstructions require more targeted monitoring in wells and polders and inclusion of emission-based source terms to improve calibration and reduce uncertainty. The Chemours case highlights the need for stronger monitoring, and the developed approach can support broader PFAS management in riverbank-filtration systems.

This study explores the feasibility of using CCTV cameras toestimate Total SuspendedSediment(TSS) concentrations in the Khlong Suan Mak River, located in Kamphaeng Phet Province, Thailand. Climate change has led to prolonged dry spells and intense precipitation in Thailand, causing significant sediment transport and deposition issues in river systems and water infrastructure. Traditional methods for monitoring TSS have inherent limitations, prompting the investigation of RGB image processing as a cost-effective alternative. Time-series RGB images were captured at multiple weir locations along the river, and pixel intensities were analyzed to estimate sediment concentrations. Synoptic and spatial sediment sampling estab lished baseline TSS values and sediment grain size distributions. Additionally, bathymetric profiles of three major reservoirs were developed to assess sediment accumulation patterns. Preliminary results demonstrated correlations between RGB intensity values, particularly in the green channel, and TSS under controlled conditions. However, field applications faced challenges due to fluctuating light intensities and minimal variability in TSS concentrations. These findings suggest that while RGB imaging shows promise for sediment monitoring, improvements in camera positioning and lighting control are essential. Future research could enhance accuracyby integrating advanced imaging technologies with this approach