We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Framework for the probabilistic modelling of mangrove ecosystem services – a Caribbean perspective
Summary
Mangrove forests can trap harmful pollutants like microplastics, heavy metals, and farm chemical runoff in their sediment, potentially keeping them out of the water we and marine life depend on. This study builds a mathematical modeling tool to better predict how well mangroves capture these contaminants, accounting for the messy, unpredictable nature of real coastal environments rather than relying on oversimplified estimates. While this is a modeling framework (tested with simulated data, not real-world measurements yet), it's a step toward smarter decisions about protecting and restoring mangroves as a natural defense against pollution.
Nature-based Solutions (NbS) have emerged as an increasingly important approach for addressing interconnected environmental and societal challenges, and mangrove ecosystems occupy a distinct position as they overlap with several NbS concepts simultaneously. However, NbS effectiveness is not easily evaluated using simplified metrics, especially within dynamic coastal ecosystems. This functional capacity is determined through complex interactions amongst structural features of the mangrove forest including species composition, hydrology, geomorphology, sediment processes and disturbance regimes that are rarely captured by deterministic approaches. Focusing on the use of mangroves for contaminant removal within the Caribbean context, of primary concern are microplastics, heavy metals and agricultural waste products which tend to be associated with mangrove sediments. This study therefore presents a probabilistic framework for modelling short-term sediment trapping as a proxy for the removal of sediment-associated contaminants from the water within mangrove ecosystems in this period. The framework is made up of three work stages and accounts for the variability introduced by ecological characteristics, hydrodynamic conditions, sediment composition and environmental drivers. To illustrate a practical application of the framework, the Monte Carlo strategy is implemented for a synthetic case study, with input random parameter values assumed to be normally distributed and deterministic variables assigned suitable values. The simulation provided the bed level change over a 1-hour period, as a random variable, where the generated probability distributions can be used to estimate net accretion which can serve as an indicator for the associated contaminant trapping potential. Other useful random outputs include the hydrodynamic forces, such as waves and currents within the mangrove forest, depicted as probability distributions. Statistical parameters of standard deviation and Standard Error of the Mean (SEM) were used to capture the uncertainty of the output. The SEM reduced considerably with increasing number of Monte Carlo runs; reducing to 0.0002 and 1.20E-10 for the wave height and bed-level change respectively at 1,000,000 runs. This illustrates that the uncertainty of the output decreases with increased sampling from the input distributions but can be balanced in practical cases for efficiency. The outcomes support evidence-based approaches for evaluating mangrove-specific NbS strategies for contaminant removal and management.