Remote sensing of ferric iron in inland surface waters

Iron concentrations in inland waters play an important role in nutrient cycling and water quality, particularly through their interaction with phosphorus, a key driver of eutrophication. Both excessive and insufficient ferric iron (Fe³⁺) levels can disrupt aquatic ecosystems. Insufficient Fe³⁺ availability hampers primary productivity, nutrient cycling, and ecosystem structure. Conversely, elevated levels of Fe³⁺ in water can pose risks to human and ecosystem health.

In Dutch agricultural-dominated lowland catchments, ferric iron-bound phosphorus is the main form of phosphorus in suspended particulate matter, potentially driving rapid transformation of dissolved phosphorus in groundwater to phosphorus in surface water. Groundwater seepage, rich in ferrous iron (Fe²⁺), further contributes to these dynamics, with Fe²⁺ oxidizing to Fe³⁺ upon exposure to oxygen, forming hydroxides that bind phosphorus. Seasonal hydrological changes also influence these interactions, with distinct red colouring observed in Dutch waters during winter attributed to iron oxidation under reduced biological activity.

A novel method using Sentinel-2 MSI data and machine learning has been developed to estimate and monitor the optically active Fe³⁺ concentration levels across Dutch surface waters in autumn and winter with a high spatial resolution (10 m). The model incorporates predictors including spectral band ratios, spectral band slopes, spectral band derivatives, and environmental variables such as air temperature and cumulative rainfall, derived from in situ data. It was trained on ~2,000 in situ iron measurements collected between 2015 and 2023.

Until now, research on Fe³⁺ in water using satellite data has been limited. This study provides a detailed spatiotemporal perspective on iron dynamics in the Netherlands, advancing the monitoring and management possibilities of water quality and ecosystem health.