Real-Time Detection of Microplastics Using an AI Camera.

Microplastics (MPs, size ≤ 5 mm) have emerged as a significant worldwide concern, threatening marine and freshwater ecosystems, and the lack of MP detection technologies is notable. The main goal of this research is the development of a camera sensor for the detection of MPs and measuring their size and velocity while in motion. This study introduces a novel methodology involving computer vision and artificial intelligence (AI) for the detection of MPs. Three different camera systems, including fixed-focus 2D and autofocus (2D and 3D), were implemented and compared. A YOLOv5-based object detection model was used to detect MPs in the captured image. DeepSORT was then implemented for tracking MPs through consecutive images. In real-time testing in a laboratory flume setting, the precision in MP counting was found to be 97%, and during field testing in a local river, the precision was 96%. This study provides foundational insights into utilizing AI for detecting MPs in different environmental settings, contributing to more effective efforts and strategies for managing and mitigating MP pollution.

Settling velocity of microplastics in turbulent open-channel flow.

The settling behavior of microplastics (MPs) plays a pivotal role in their transport and fate in aquatic environments, but the dominant mechanisms and physics governing the settling of MPs in rivers remain poorly understood. To gain mechanistic insights into the velocity lag of MPs in an open-channel flume under different turbulent flow conditions, an experimental study was conducted using three types of MPs: polystyrene, cellulose acetate, and acrylic, of sphere-shaped particles with diameters ranging from 1 mm to 5 mm. A particle tracking technique was employed to record and analyze the MPs velocity within turbulent flows. The results showed a variation in the vertical settling velocity of MPs ωMP ranging from -26 % to +16 %, when compared to their counterparts in still water (ωs). A new formula for the drag coefficient (Cd) of MP particles was developed by introducing the suspension number (u∗/ωs). The developed Cd formula was used to calculate the resultant velocity lag VMP, with a mean relative error of 16 % compared with the measured values. Further, the study highlighted that the MPs with large Stokes numbers are mainly driven by their own inertia and turbulence has less influence on their settling behavior. This study is crucial for understanding the settling behavior of MPs in turbulent flows and developing their transport and fate models for MPs in riverine systems.

Settling behaviors of microplastic disks in acceleration fall.

The settling of microplastics (MPs) in the initial acceleration fall stage, i.e., before reaching the terminal settling velocity, has not been investigated, which is however important for understanding MP transport and fate. MP disks sized 3-5 mm, of three shapes and made of three polymers (1.038-1.343 g/cm3) were examined. Five release ways and three release angles (0°, 45°, 90°) were used. MP disks with the release angle of 0° start to zigzag immediately after the release, while the MP disks with the release angles of 45° and 90° first adjust to a horizontal position and then zigzag. The adjustment distances in the vertical and horizontal directions, as well as the maximum vertical settling velocity, are influenced by MP density, size, release angle and release way. The detailed settling trajectory and velocity were also analyzed. Finally, the time-changing drag coefficient of MP disks was examined and discussed.

Distribution and risk assessment of microplastic pollution in a rural river system near a wastewater treatment plant, hydro-dam, and river confluence.

Rivers are the natural drainage system, transporting anthropogenic wastes and pollution, including microplastics (plastic < 5 mm). In a riverine system, microplastics can enter from different sources, and have spatial variance in concentration, physical and chemical properties, and imposed risk to the ecosystem. This pilot study presents an examination of microplastics in water and sediment samples using a single sample collection from the rural Raquette River, NY to evaluate a hypothesis that distinct locations of the river, such as downstream of a wastewater treatment plant, upstream of a hydro-dam, and river confluence, may be locations of higher microplastics concentration. In general, our results revealed the presence of high microplastic concentrations downstream of the wastewater treatment plant (in sediments), upstream of the hydro dam (both water and sediment), and in the river confluence (water sample), compared to other study sites. Moreover, the risk assessment indicates that even in a rural river with most of its drainage basin comprising forested and agricultural land, water, and sediment samples at all three locations are polluted with microplastics (pollution load index, PLI > 1; PLIzone = 1.87 and 1.68 for water and sediment samples respectively), with risk categories between Levels I and IV ("minor" to "danger"). Overall, the river stands in a "considerable" risk category (PRIzone = 134 and 113 for water and sediment samples respectively). The overall objective of this pilot study was to evaluate our hypothesis and advance our understanding of microplastic dynamics in rural river systems, elucidating their introduction from a point source (wastewater treatment plant), transit through an impediment (hydro-dam), and release into a vital transboundary river (confluence of Raquette-St. Lawrence Rivers).

The effects of large roughness elements on the in-stream transport and retention of polystyrene microplastics.

The mechanisms controlling transport and retention of microplastics (MPs) in riverine systems are not understood well. We investigated the impact of large roughness elements (LREs) on in-stream transport and retention of the ubiquitous polystyrene-microplastics (PS-MPs). Scaled experiments were conducted with and without LREs under various shear Reynolds numbers (Re*) in an ecohydraulics flume. Our results, for the first time, demonstrated a clear dependence of the MPs' velocity on Re* in LREs-dominated channel. Two distinct regimes and thresholds were identified: lower Re* (≤ 15,000) regime corresponding to higher velocities of MPs ([Formula: see text]> 0.45), and higher Re* (> 15,000) to lower [Formula: see text]< 0.45). The presence and higher density of LREs increased Re*, decreased [Formula: see text], and enhanced the PS-MPs capture. The LREs-generated turbulence kinetic energy (TKE) was found to be a good predictor of PS-MPs transport and retention rates, indicating the effectiveness of LREs in retaining PS-MPs in streams and rivers.

Settling behaviors of microplastic disks in water.

Microplastic (MP) disks have not been studied for settling behaviors in aquatic environments, which affects the transport and fate of MPs. Therefore, settling experiments were conducted on MP disks of three shapes and four common-seen materials. Lighter MP disks (with density ρs = 1.038 g/cm3 and length l ≤ 5 mm) followed rectilinear vertical trajectories, while heavier MP disks (ρs = 1.161-1.343 g/cm3 and l = 5 mm) followed zigzag trajectories with oscillations and rotations. The mean terminal settling velocities of MP disks were 19.6-48.8 mm/s. Instantaneous settling velocities of heavier MP disks fluctuated. Existing formulas could not accurately predict the settling velocity of MP disks; thus, a new model was proposed with an error of 15.5 %. Finally, the Red - I* diagram (Red is the disk Reynolds number and I* is the dimensionless moment of inertia) was extended for MP disks to predict settling trajectories.

An experimental study: effects of boulder placement on hydraulic metrics of instream habitat complexity.

Boulder placement is a common method to restore degraded instream habitats by enhancing habitat complexity. This experimental study is the foremost attempt to systematically investigate the influence of rock-ramp boulder placement with varying boulder concentration and flow rate on habitat hydraulic complexity metrics, including the kinetic energy gradient and modified recirculation metrics. By adding boulders to a reach, the modified recirculation metric increased by one order of magnitude for all boulder concentrations. Based on the studied metrics, boulder placement with the highest boulder concentration (λ = 8.3%) resulted in the greatest habitat hydraulic complexity. A set of relationships of moderate strength were proposed to predict the metrics in reaches with boulders by having information about only boulder concentration, habitat characteristic size, and reach-averaged flow characteristics. Based on the available data from the literature, boulder placement especially at higher concentrations may provide suitable habitats for several riverine fish species. Further studies are needed to establish a reliable linkage between the metrics and instream species, to test a wider variety of parameters for verifying and improving the range of applicability of the proposed relationships, and to find the structural configuration at which the habitat complexity is maximized or optimized for a certain species.