Introduction

The occurrence of microplastics (MPs), usually defined as particle sizes less than 5 mm in length (Frias and Nash 2019; Dahms et al. 2020), in riverine ecosystems represents an ecological risk of concern. Laboratory and field-based studies have shown effects of MP on different riverine organisms, including invertebrates (Welden and Cowie 2016; Scherer et al. 2017; Akindele et al. 2019), fish (McNeish et al. 2018; Kühn et al. 2020; Sequeira et al. 2020) and birds (Lavers et al. 2019). Commonly reported effects of MP include gut filling, abrasion, entanglement, and release of potentially toxic chemicals into the environment (Ma et al. 2019; López-Rojo et al. 2020; Issac and Kandasubramanian 2021). Studies of MP are common for the marine environment (Berlino et al. 2021; Petersen and Hubbart 2021; Tang et al. 2021), but in recent years, there has been a growing interest in MP in freshwater ecosystems, particularly in riverine systems (Windsor et al. 2019; Frank et al. 2021; Yang et al. 2021). Field-based studies have demonstrated that MPs severely influence freshwater systems (e.g. rivers) and that rivers are essential sinks and regulators of MP transport to marine systems (Klein et al. 2015; Mani et al. 2015; Hoellein et al. 2019). Although lentic and marine habitats are the most probable ultimate sinks of MP (Berlino et al. 2021; Tang et al. 2021), rivers may act as conduits and MPs accumulators (Hoellein et al. 2019). The significance of the riverine ecosystem in MPs pollution studies is highlighted by its role in the regulation of MPs transport to other aquatic systems such as lakes and oceans, though the proportion of input from rivers relative to other sources remains unknown.

Microplastics in rivers originate from primary and secondary sources. Primary sources comprise manufactured plastic products such as scrubbers in cleaning and cosmetic products, as well as manufactured industrial pellets that serve as precursors for plastic production (Hidalgo-Ruz et al. 2012). Manufactured pellets may be especially common in the environment near plastic-processing plants whereas scrubbers may be present in industrial and domestic wastewater which is usually emptied into rivers (Eerkes-Medrano et al. 2015). Secondary sources of MPs consist of fibres or fragments of plastic resulting from the breakdown of larger plastic products (Hidalgo-Ruz et al. 2012; Eerkes-Medrano et al. 2015). Rivers are exposed to the same sources of MPs as marine environments, although secondary sources are believed to be the main origin of most MPs in marine environments (Hidalgo-Ruz et al. 2012; McCormick et al. 2016).

Riverine systems are highly important in the studies associated with MPs pollution. Rivers are highly susceptible to inputs from primary sources because of closer proximity to point sources of MPs pollution, and they have relatively little water volume for MPs dilution, which suggests they have high concentrations (McCormick et al. 2016). High MPs concentrations have been reported in riverine sediments and surface waters (Klein et al. 2015; Mani et al. 2015; Choong et al. 2021; Osorio et al. 2021) which is a concern because of the potential for high inputs into marine systems, trophic transfer among biota, and the potential impact on human health as human populations depend substantially on riverine resources. The interaction of human populations with riverine systems, and the potential for the transfer of MPs into humans is a critical area of research.

Further, riverine systems show a marked difference from their marine counterparts. Riverine systems are hierarchically structured and characterised by different flow environments and hydraulic attributes that impact the distribution of particles distinctively. Also, rivers are relatively small in size with spatial and temporal differences in the mixing and transport of particles by physical forces (Eerkes-Medrano et al. 2015). The dissimilarities between riverine and marine systems may lead to differences in the type of MPs present (Eerkes-Medrano et al. 2015) and the configuration of MPs concentration. For instance, rivers may show a predictable pattern in MPs characteristics (e.g. size, shape, relative abundance) based on anthropogenic activities and land use type adjacent to the river (Eerkes-Medrano et al. 2015). Rivers may also show a predictable pattern in MPs transportation and deposition based on MPs physical characteristics and the hydrodynamic attributes (e.g. flow velocity, water depth) in different flow environments (e.g. pool, run, rapid, backwater etc.). Therefore, a clear understanding of the accumulation sites, distribution, and transport dynamics of MPs across riverine systems is critical for unravelling impacts on riverine ecosystem processes, biota, and potential inputs into marine systems.

Microplastic transport and concentration in riverine ecosystems are influenced by river hydro-geomorphology (Kiss et al. 2021) and the physical characteristics of MPs such as density, shape, and size (Kumar et al. 2021), although knowledge on how different flow environments and morphological units could shape the distribution and concentration of MPs along a river is scant. Hydro-geomorphology refers to the hydrological and geomorphological processes and forms that occur within catchments and their river systems (Gurnell et al. 2016). Within individual morphological units are nested different flow compartments which have been described as hydraulic biotopes (Wadeson and Rowntree 1998; Thomson et al. 2001) or morphohydraulic units (Lehotský, 2004) or simply hydraulic units (Thomson et al. 2001). The term hydraulic biotope was preferred in this paper rather than hydraulic unit or morphohydraulic unit because it brings into sharp focus the interaction between biology and hydraulics. A hydraulic biotope is a spatially distinct instream flow environment (e.g. run, pool, glide) characterised by specific hydraulic attributes that provides the abiotic environment in which species assemblages or communities live (Wadeson and Rowntree 1998).

Hydraulic biotopes (e.g. pool, run, rapid, etc.) are characterised by distinct hydraulic patterns that result from the interaction between flow, substrate, and channel morphology (Wadeson and Rowntree 1998), which are important factors that impact the transport dynamics of MPs in riverine systems (Kiss et al. 2021; Kumar et al. 2021; Newbould et al. 2021). The hierarchical structure of rivers suggests that the influence of different hydro-geomorphological processes and forms are most impactful at the lower spatial hierarchy (Thomson et al. 2001; Lamouroux et al. 2004; Gurnell et al. 2016) such as the hydraulic biotope. Thus, the hydraulic biotope which occur at a fine spatial scale in the river hierarchy (Wadeson and Rowntree 1998; Thomson et al. 2001) can be used to unravel the potential impact of different hydro-geomorphological processes and forms on the distribution and concentration of MPs and the potential exposure of riverine biota.

Furthermore, the hydraulic attributes (flow-related abiotic factors, e.g. substrate type, depth, flow velocity) in each hydraulic biotope (Wadeson and Rowntree 1998; Biswas and Chandra Das 2016) impact on the deposition, suspension, and remobilisation of MPs (Kumar et al. 2021; Newbould et al. 2021). For example, dimensionless hydraulic indices developed through the integration of depth, velocity, and bed roughness can provide insight into the prevailing hydrodynamic conditions in each distinct biotope (Jowett and Richardson 1990; Jowett et al. 1991; Wadeson 1994; Wadeson and Rowntree 1998), which in turn, can inform our understanding of how hydraulic biotopes may influence the behaviour of MPs and the instream factors impacting their concentrations in different morphological units and landforms in river systems.

At the reach scale of river spatial hierarchy, the hydraulic biotopes play a critical role in understanding the transport dynamics of MPs in riverine systems because of the integration of flow velocity, depth, and bed roughness. In addition, studies have shown that organisms tend to show differential preferences for different hydraulic biotopes (Jowett and Richardson 1990; Jowett et al. 1991). Thus, the hydraulic biotopes may mediate organismal interactions with MPs and determine the residence time and eventual fate of MPs as well as exposure of organisms to MPs. Despite the role of hydraulic biotopes in shaping biological assemblage distribution in streams, and the transport dynamics of MPs, not much consideration has been given to it in MPs research.

There is a growing body of studies that have reported the linkage between hydrology, geomorphology and MPs transports in rivers (Klein et al. 2015; Mani et al. 2015; Zhang 2017; Rodrigues et al. 2018; Campanale et al. 2020; Defontaine et al. 2020; Kumar et al. 2021; Lu et al. 2023), but none of these addresses the question of the role of hydraulic biotopes, geomorphological, biological and allochthonous factors on MP transports at the reach scale, and the ecological significance thereof. Knowledge of the influences of the varieties of hydraulic biotopes on MPs hydrodynamic behaviour and transport in rivers is therefore limited. Further, at different flow regimes, the transport dynamics and concentration differences of MPs in different hydraulic biotopes are unclear. Besides, the hierarchical structure of rivers, the heterogeneous nature of MPs and the diverse concentrations of MPs reported for the riverine network (Klein et al. 2015; Mani et al. 2015; Rodrigues et al. 2018; Campanale et al. 2020) suggest that further research and innovation is required to understand MPs transport in the river ecosystem at the reach scale. The current review analyses environmental factors that impact MPs transport in rivers with emphasis on the influences of hydro-geomorphological factors. New insights on how the concept of hydraulic biotopes can be applied in unravelling MPs transport behaviour and quantification in rivers is provided, as well as a synthesis of the ecological implications of these new insights for MPs studies in rivers. It is argued that explicit consideration of the role of hydraulic biotopes in MPs research can have far-reaching implications in terms of deciding where to collect samples within a river system and can predict potential distribution and concentrations of MPs. This paper fills an important knowledge gap by presenting a review of the role of hydro-geomorphological factors in the transport dynamics of MPs in a river system. The interplay between MPs characteristics and hydraulics is also reviewed in relation to the potential distribution of MPs in river systems.

The paper is divided into nine sections. Section one provides a background to the review. In Section two, the notion of rivers as physical habitats is explored, while in Section three, we review the transport of MPs in rivers. In Sections four, five, six and seven we assess factors that impact MPs transport in riverine systems. In Sections eight and nine, the ecological implications of the paper for MPs research are synthesised.

Rivers as physical habitats

Rivers are complex biophysical and hydrogeochemical systems (Dollar et al. 2007). Viewed spatially, rivers can be construed as hierarchical systems with smaller units (e.g. hydraulic biotopes) nested within larger ones (e.g. geomorphic units) (Frissell et al. 1986; Thomson et al. 2001; Kondolf et al. 2016a, b). Drawing from the river morphology hierarchical classification by Lehotský (2004), rivers are characterised by 7 spatial levels, i.e. catchment-drainage network, zone, segment, channel-floodplain unit, river reach, morphological unit and morphohydraulic unit. (i.e. hydraulic biotopes). The hydraulic biotope constitutes the smallest spatial level in the river hierarchy and is nested in the larger morphological spatial hierarchy (Wadeson 1994; Thomson et al. 2001; Lehotský, 2004; Gurnell et al. 2016). Generally, the smaller spatial levels are constrained by the larger levels (Thomson et al. 2001). The hydraulic biotope represents different instream flow environments with distinct hydraulic patterns that result from the interaction between flow, substrate, and channel morphology (Wadeson and Rowntree 1998; Thomson et al. 2001)—three factors that impact the distribution of MPs in river ecosystems. It is important therefore, to unravel how the hydraulic biotope impacts the patchy distribution of MPs and contribute to our understanding of MPs transport dynamics in rivers.

At the reach scale, the variety of hydraulic biotopes present in a morphological unit and their distinct hydraulic patterns give a picture of the hydrodynamic conditions of the river environment (Jowett 1993; Wadeson 1994, 1996; Rowntree and Wadeson, 1999). The distinct hydraulic patterns in hydraulic biotopes, which are a function of the interaction between flow, substrate, and channel morphology, together with the physical properties of MPs, determine the hydrodynamic behaviour of MPs (i.e. suspension, aggregation, remobilisation, etc.). The hydrodynamic behaviour of MPs eventually determines the fate of MPs in the river systems (Yan et al. 2021a). Therefore, at the reach scale, the hydraulic biotope is a critical consideration for understanding the behaviour and eventual fate of MPs.

Dimensionless indices such as velocity/depth ratio, Froude number, Reynolds number, shear velocity, and "roughness" Reynolds number have been used to distinguish and describe the mean motion of flow in the water column and near-bed hydraulics in hydraulic biotopes (Jowett 1993; Rowntree et al. 1999). The indices thus have several implications for MPs behaviour (e.g. deposition, suspension, remobilisation, or horizontal transport, etc.). For instance, in pool biotopes with sub-critical tranquil flow, the movement of suspended MPs and the remobilisation of settled MPs is likely to be slow compared to rapids with super-critical turbulent flows where the rate of remobilisation and translocation of MPs are likely to be higher. Thus, the patchy concentrations and behaviour of MPs within river reaches may be driven more by these biotopes than by input from a potential source.

Understanding the hydrodynamic conditions in hydraulic biotopes is useful for the study of MPs in riverine systems in several ways. First, at the reach scale, disparity in the concentrations and density of MPs that exist within rivers, or different river morphological units can be interpreted based on the hydrodynamic characteristics of the hydraulic biotopes sampled within a particular morphological unit. Second, the hydraulic biotope offers an ecologically useful way of comparing sites in terms of their MPs concentration and transport dynamics. This way, studies can be standardised in terms of collecting technique, sampling sites, and measurement of hydraulic indices for inter-laboratory calibration and study comparison. Third, the hydraulic biotopes present the most straightforward scale within riverine ecosystems where the impact of hydro-morphological variations on the occurrence and transport of MPs can easily be understood and related to the biotic community. To better understand the way in which hydraulic characteristics impact MPs transport, the next section is devoted to a review of MPs transport in riverine systems.

Microplastics transport in rivers

Three forms of transport patterns can be deduced for MPs in riverine systems. These are horizontal transport, vertical transport, and lateral transport. Horizontal transport occurs when MPs follows the natural upstream and downstream river flow, whereas vertical transport is driven primarily by gravity, density and turbulence, allowing settling or remobilisation of settled MPs. Lateral transport is driven primarily by allochthonous factors such as wind, runoffs, or discharges from the adjacent landscape.

Several important biophysical and MPs characteristics play an important role in influencing each of the transport forms. For example, horizontal transport is influenced by hydraulic factors, such as flow velocity, turbulence (He et al. 2020); climatic factors, such as rainfall (Yan et al. 2021a); biological factors, such as vegetation patchiness; MPs characteristics, such as size and density, and river anthropogenic alterations, such as dams. Regarding vertical transport for instance, polymer density and size play an important role. Often polymers denser than water tend to sink to the river bottom and may remain settled or be remobilised into suspension by hydraulic factors such as flow velocity. Ballent et al. (2012) demonstrated that plastic density markedly influences vertical displacement, followed by size and flatness. Besides particle density and size, particle shape may also play a part in the sinking of MPs (Kowalski et al. 2016). Biofilm formation on MPs usually alters their density. Owing to biofilms, MPs usually considered less dense than water may become denser and then sink to the riverbed. The development of biofilms tends to increase the efficiency of forming aggregates of more prominent and denser particles (Miller and Orbock Miller 2020), thus impacting vertical movement. Although buoyant, turbulent forces also transport MPs vertically in the water column by Zhang (2017); vertical movement generally favours denser MPs in biotopes with sub-critical to critical tranquil flow where both the level of disturbance (i.e. turbulence) and speed of flow is markedly reduced.

Microplastics found in rivers originate from a broad range of terrestrial sources, including land cover and several anthropogenic activities in proximity to freshwater bodies. Terrestrial sources of MPs could be both point and non-point sources (Kataoka et al. 2019a). In particular, population density and urban ratio have been correlated with MPs abundance in rivers (Kataoka et al. 2019b; Li et al. 2021). Microplastics from land-based sources are transported to freshwater environments by several pathways such as precipitation, runoff (Talbot and Chang 2021), rainwater in roadside gutters, wind, and storms (Kataoka et al. 2019b).

Plastics littered on the open landfill near rivers may be fragmented into MPs and released into rivers by rainfall and strong wind (Kataoka et al. 2019b). Once in the aquatic environment, MPs are subjected to hydrodynamic processes which may influence their accumulation or deposition (Kumar et al. 2021; Yan et al. 2021b). However, the timing and volume of precipitation and runoff, the intensity of wind that drives this lateral movement of MPs, the impact of windbreakers and other terrestrial debris barriers, and how they influence the eventual input into freshwater systems need to be clearly understood. In the next section, we review in detail the key factors that influence MPs transport in riverine systems. First, we review river-related factors, then MPs characteristics, followed by allochthonous factors, and lastly, biological factors.

River-related factors influencing MP transport dynamics

Autochthonous river factors that may impact the transport of MPs in riverine systems include channel morphology (e.g. channel width, channel slope, watershed pattern, substrate character including bed roughness, and clast size), hydraulic factors (e.g. flow velocity, water depth), discharge, and stream power (Kumar et al. 2021; Newbould et al. 2021). These factors and their potential influence on MPs transport are reviewed below.

Channel morphology and hydraulic factors

The behaviours of MPs (e.g. deposition, suspension, remobilisation, aggregation, etc.) are impacted by different hydraulic patterns and channel morphology, especially within the flow-dependent hydraulic biotopes. For instance, Miller and Orbock (2020) stated that MPs concentrations, transport and storage in river sediments are highly susceptible to changes in flow conditions over a range of temporal scales. Wang et al. (2021) observed a positive correlation between river channel width and MPs larger than 2 mm, and a negative correlation with smaller MPs. Little is known about the level of resistance posed by different substrate types to flow in relation to MPs transport. The amount of shear stress that must be surpassed within the terminal sub-layer in different biotopes to remobilise settled MPs also remains an important area of research for a better understanding of MPs hydrodynamic behaviour in riverine systems.

Rowntree et al. (1999) suggested that flow velocity and depth are two hydraulic factors that may act independently or in combination through some hydraulic indices which describe either the mean flow (average conditions in the water profile) or near-bed conditions. These hydraulic indices describe both the mean flow motion in the whole water column and the microflow environment in which many aquatic communities live either on or near the channel bed (Wadeson and Rowntree 1998). The velocity/depth ratio (mean), Reynolds number (Re), and the Froude number (Fr) describe the mean flow in the water column. In contrast, near-bed hydraulic conditions have been described by hydraulic indices such as the shear velocity (U*) and the ‘roughness’ Reynolds number (Re*) (Wadeson 1994; Wadeson and Rowntree 1998). The role of these indices in describing the situation in terms of flow hydraulics is critical to understanding the hydrodynamic behaviour of MPs in different flow environments because the inherent density of MPs particles and the hydrodynamic conditions of the river govern the trajectory pathways of MPs particles (Kumar et al. 2021).

Kumar et al. (2021) noted that, depending upon the Re, either laminar (Re < 1) or turbulent (Re > 100) flow can impact the flow dynamics of MPs concerning the river depth profile; laminar regime at Re < 1, turbulent regime at 103 < Re < 105, and the transitional regime, in-between (at 1 < Re < 103). At low Re, the water is highly viscous, mixing is impeded, and the transport of materials is slowed (Rowntree and Wadeson, 1999). The implication is that high-density MPs particles can dominate the river bottom in laminar flow conditions, while low-density MPs particles are in suspension. The low-density MPs particles travel for long distances in turbulent flow conditions, while the high-density MPs can either be in suspension or settle on the river bottom after a certain translocation distance (Kumar et al. 2021).

The Fr is a dimensionless number used to differentiate between sub-critical flows (where Fr < 1), that is, slow or tranquil flow, and super-critical flows (Fr > 1), that is, fast or rapid flow. In sub-critical flows, the role played by gravity forces is more pronounced, so that flow has a low velocity relative to depth (Rowntree and Wadeson, 1999; Bravard and Petit 2009). The behaviour and movement of MPs is impacted depending on the flow type, that is, whether it is either sub-critical or super-critical. For example, sub-critical flow (Fr < 1) can potentially increase the sinking and settlement of MPs. In contrast, super-critical flows, that is, rapid flow and flooding events (Fr > 1) can potentially increase the transport and aggregation of MPs.

The roughness of the substrate material also influences flow. The height of the roughness elements relative to the thickness of the laminar sub-layer (i.e. the region close to the bed where flow is entirely laminar) is an essential determinant of flow conditions near the bed (Rowntree and Wadeson, 1999). If roughness height extends above the laminar sub-layer, it affects the surface flow. River roughness depends mainly on the nature of bed material, on vegetation that can clutter the channel, on bedform succession, and the presence of sinuosity (Bravard and Petit 2009). Roughness generally varies between 0.020 and 0.100 but can exceed 0.150 and even reach a value of 0.42 for a small rivulet with aquatic vegetation at a low water stage (Bravard and Petit 2009). River roughness impacts stream power as velocity decreases with increasing bed roughness and decreasing slope (Rowntree and Wadeson, 1999). The Re*, therefore, combines the effects of velocity and substrate type. Thus, flow near the riverbed will be disturbed either when the roughness elements increase in height or when the velocity increases, causing the laminar sub-layer to become smaller than the height of the projections (Rowntree and Wadeson, 1999). Low Re* will, therefore, potentially increase the transport of more MPs than high Re*. Also, MPs transport will receive the least resistance in hydraulically smooth surfaces (Re* < 5), and the most resistance in hydraulically rough surfaces (Re* > 70), and transition at 5 < Re*, < 70.

Bed shear stress is a force per unit area of the bed (in N m−2) and increases with flow depth and steepness of the channel (Biswas and Chandra Das 2016). The coarse particles lying on riverbeds are set into motion during floods when critical shear stress is passed (Bravard and Petit 2009). Thus, laminar and turbulent flow can induce critical shear stress, in which uneven riverine bottom surfaces can play a significant role in the resuspension of MPs (Kumar et al. 2021). We thus suggest the integration of hydrodynamic conditions indicated by the different hydrodynamic indices in river-related MPs field studies to better understand MPs distribution patterns and concentration configurations in rivers.

Stream power

The concept of stream power was first used in fluvial geomorphology to study sediment transport in river channels. It was later applied for other purposes such as bedload, suspended load, channel instability, bank erosion, thresholds in channel patterns, and flood plain dynamics (Kondolf et al. 2016b). Stream power (Ω) is the rate at which a stream performs its geomorphologic work and is a good measure of its capacity to erode and transport materials (Biswas and Chandra Das 2016). It determines the capacity of a given flow to entrain and transport sediment; thus, stream power increases with increasing gradient and flow discharge (Rowntree and Wadeson, 1999).

Stream power should impact the movement of high-density, bottom-settling MPs, including biofouled MPs, homoaggregates, and heteroaggregates. Discharge, flow depth, and channel slope determine the river's stream power and shear stress (Biswas and Chandra Das 2016). Stream power is thus an important variable that needs to be considered in MPs transport. Particularly, insights from research in the field of sediment transport can prove useful in our understanding of the interaction between stream power and MPs transport.

Stream power is measured in watts per unit channel length, usually Wm−1, and is calculated using the formula:

$$\Omega = \rho \mathrm{gQS}$$

where Ω is stream power per unit of stream length in units of kg-m/s3; ρ is water density, that is, 1000 kg/m3; g is the acceleration due to gravity, that is, 9.81 m/s2; Q is discharge, and S is channel slope, that is, m/m.

Characteristics of microplastics that influence their transport

Microplastic particles possess intrinsic properties that impact their transport and distribution. In this section, we review these properties and how they may impact MPs transport and distribution in riverine systems.

Density

The density of MPs varies considerably, depending on the polymer types and manufacturing process (Zhang 2017). Particle density can influence the deposition and mobility of MPs in riverine ecosystems (Kumar et al. 2021). Microplastic in the density range more diminutive than the river water floats or remains in suspension. In contrast, the higher density MPs tends to deposit over the riverbed (Kumar et al. 2021). Because of aggregation, mechanical breakdown, biofouling, photochemical degradation, and flocculation mechanisms, the density of plastic particles might vary with residence time in the riverine ecosystem (Kumar et al. 2021; Zhang 2017), thus impacting their eventual fate.

Regarding aggregation, Zhang (2017) noted that aggregation with organic and inorganic particles could increase the size and density of MPs and cause rapid deposition onto benthic sediments. Nguyen et al. (2020) also demonstrated the formation of microbial-associated MPs aggregates possibly caused by cell colonisation. Hoellein et al. (2019) observed that the denser fragments could easily be retained and may resist remobilisation, depending on the degree to which they are attached or buried in bed sediments. Many low-density polymers remain in suspension due to buoyancy. Overall, even though density plays an important role in the transport and distribution of MPs, density is usually altered by several factors such as (i) MP residence time, (ii) MP-microbial interaction, (iii) biofouling, (iv) MP-organic chemical interactions, (v) aggregation and breakdown. A combination of these factors implies that predicting the potential distribution and transport of MPs based on their density is complex and not straightforward. How these factors interact collectively to influence MPs distribution and transport remains an important area of research.

Shape and size of microplastics

Plastic particles in the riverine ecosystem exist in various shapes (e.g. fibres, pellets, fragments, films, foams, microbeads, and granules) (Kumar et al. 2021). Generally, low-density fibres and films have high buoyancy and low settling velocity. Fibres can remain in suspension for longer than fragments and spherical beads in the water column (Kumar et al. 2021). Hoellein et al. (2019) reported that fragments have high deposition velocity, followed by fibres and then pellets; therefore, fibres and pellets tend to travel longer than other particle shapes. Deposition velocity is also impacted by particle size of certain MPs shapes with the effect of particle shape more pronounced for larger particles (Khatmullina and Isachenko 2017).

Smaller particle sizes with a high surface-area-to-volume ratio are more prone to aggregation than are larger MPs particles (Wang et al. 2021; Yan et al. 2021b). Aggregation generally increases their density and impact on vertical transport. The effect of surface-area-to-volume ratio is commonest among fibres, films, and foams (Kumar et al. 2021).

Allochthonous factors

The allochthonous factors which impact MPs transport that are discussed in this section include wind and precipitation and point and non-point discharges.

Wind and precipitation

Wind and precipitation impact the lateral transport of MPs (Han et al. 2022; Rezaei et al. 2019). Microplastics get into land by several means; (i) direct littering and inefficient waste management; ii) plastic use for mulching and wrapping in modern-day agriculture; (iii) settled MPs introduced along with sewage sludge into agricultural lands as fertilisers. Once on land, their characteristic light weight makes it possible for MPs to be easily transported from adjacent land into surrounding rivers by either wind or rainfall-induced surface runoff. This movement, especially during soil erosion or surface runoff, follows the hydrological and sediment transport pathways (Lwanga et al. 2022) impacting markedly on lateral transport into rivers. For instance, Cheung et al. (2019) recorded nearly double the number of MPs found on coastal sea surfaces as the same area in an urban river following a three-day rainfall event in Hong Kong.

Agricultural drainage and runoff from farmlands and other forms of land-use types can result in input of MPs and sewage-sludge-derived fibres and microbeads (Horton and Dixon 2018). Storm drainage and urban runoff can contain MPs from degraded road paint, plastic litter in gutters, and wear from machinery (Horton and Dixon 2018) and have an impact on lateral movement. However, the effect of the interaction of MPs with soil aggregates, different rainfall scenarios (including rainfall intensity and frequency), the intensity of runoffs and how they may impact lateral transport of MPs needs to be carefully researched. The presence of windbreakers such as trees and shrubs, interception of MPs and obstruction of surface runoff and sediment delivery pathways into rivers by physical barriers such as terrestrial vegetation, depressions, bends, and uneven land surfaces, and their impact on the lateral transport of MP is also an interesting area of research.

Point and non-point discharges

Studies have often linked lateral movement of MPs to occur near densely populated or urban areas affected by point and non-point source pollution (McCormick et al. 2014; Kataoka et al. 2019a; Liu et al. 2021). Discharges from point sources (e.g. direct discharge from sewers and drains) and non-point discharges such as runoff from different land-use types (e.g. farmlands and cities) are seen as the major factors impacting the movement of MPs into rivers, but point discharges such as effluent outfalls, combined sewage overflows and storm drains which are usually isolated from densely populated areas (Horton and Dixon 2018) markedly impact the lateral transport of MPs.

As urban developments are a major sources of MPs from point and non-point discharges, it is expected that the concentration of MPs in urban rivers would be higher, compared to those in non-urban areas (McCormick et al. 2014; Kataoka et al. 2019a; Koutnik et al. 2021; Liu et al. 2021;Wang et al. 2021). The proximity of rivers to point and non-point MPs emission sources is a major factor that contributes to a wider variability in MPs concentrations in rivers (Koutnik et al. 2021). For example, Mani et al. (2015), reported a diverse concentration of MPs along and across the river Rhine in Europe, reflecting various emission sources. These concentration reflected by various emission sources can thus be patchily distributed down river length via the hydrological and hydraulics variables (Koutnik et al. 2021; Kumar et al. 2021). For instance, high flows have been reportedly linked to increased concentration of suspended MPs (Lu et al. 2023). Also, high flow velocity facilitates the transport of more MPs from source points and for longer distance (Lu et al. 2023).

The water regime and morphology (e.g. channel width) can affect the distribution of various pollutants, such as heavy metal, pesticide, and polycyclic aromatic hydrocarbons (PAHs) (Zhang 2017; Wang et al. 2021); whether these factors can affect the in situ transport patterns of MPs in rivers, remain largely unknown. Seasonality and flooding events can also potentially influence the abundances of MPs in rivers (Kataoka et al. 2019b; Xia et al. 2021). As a result of reinforced hydrodynamic conditions during rainfall, MPs deposited in sediments can remobilised and return to the overlying water, which facilitates the horizontal transport of microplastics (Yan et al. 2021a). The patchy concentration of MPs in rivers from different source points needs further investigation to unravel their fate in different flow compartments or hydraulic biotopes such as pools and runs, etc., during different flow conditions in other to detect potential sinks.

Biological factors

Microbial colonisation, ingestion by aquatic organisms, and vegetation patchiness are the biological factors reviewed in this section. Microplastics undergo biological interactions through ingestion (Zhang 2017) and biofilm formation (McCormick et al. 2014). Microplastics can readily develop surface fouling by accumulating microbial films, followed by colonisation by algae and invertebrates, affecting their size and thickness and thus their vertical and horizontal transport. McCormick et al. (2014) observed extensive microbial colonisation of MPs pellets and fragments. This colonisation may be because pellets and fragments have a large surface-area-to-volume ratio, which provides space for microbial colonisation. However, whether microbial colonisation of MPs and subsequent biofilm formation is influenced by MPs properties and their location in the water column remains an important area of research. The colonisation of MPs by micro-organisms is an important factor affecting heterogenous aggregation of MPs (Yan et al. 2021b) with implications for vertical transport of heteroaggregates.

The transport of MPs is also influenced by the grazing of aquatic organisms such as fish, tadpoles, and predatory macroinvertebrates. Studies have suggested that aquatic organisms may ingest MPs that resemble their food (Steer et al. 2017; Tien et al. 2020). The formation of biofilms on MPs surfaces has been reported to particularly increase the chances of ingestion owing to their taste, smell, and colour, which may be similar to that of plankton (Stabnikova et al. 2021). Following ingestion, MPs particles pass through the gut system intact. Eventually, these particles are egested alongside other debris, which may lead to the formation of aggregates. The formation of aggregates alters their density and ultimately impacts vertical transport.

Vegetation (both aquatic and riparian) may act as physical traps of MPs (Newbould et al. 2021) impacting horizontal transport. The impact of vegetation on MPs transport could be exacerbated when MPs interact with other barriers such as hydraulic traps, fallen logs, emergent boulders, et cetera, which may act to trap MPs for a long time. Trapped MPs may either undergo aggregation, forming homoaggregates and heteroaggregates, that could increase their density and cause them to sink to the riverbed (i.e. vertical transport) or are eventually released from the traps during higher flows or flooding, with an impact on horizontal transport. The impact of MPs aggregation on MPs transportation and concentration in rivers need to be further investigated to understand overall impact on MP sinks in rivers and discharge into marine systems (i.e. MP conduit and regulatory role of rivers).

Ecological implications

Our review has important implications for the ecological studies of MPs in riverine systems. The first implication is the potential spatial distribution of MPs in riverine systems and the design of MPs sampling protocols, that is, where and how to collect MPs samples. Microplastic abundance measurement in rivers is mostly done along a gradient of land use types, point sources of pollution, and generally along a river stretch, either to ascertain inputs from anthropogenic activities or generally to quantify abundance (Mani et al. 2015; He et al. 2020; Choong et al. 2021) without consideration for the impact of the different flow environments mostly generated by river morphology. It is clear that at the reach scale, hydraulic biotopes reflect substrate type, depth, and flow velocity that influence MPs particle and biota distribution (Jowett et al. 1991; Wadeson and Rowntree 1998; Thomson et al. 2001). Miller and Orbock (2020) stated that a critical finding of using a geomorphological approach for assessing rivers affected by hydrophobic contaminants is that sediment-associated contaminants are not randomly distributed within the channel bed but are partitioned into specific depositional features within the channel, creating variations in spatial concentrations of these contaminants between channel bed features (e.g. pools, glides, riffles) as a function of their physical attributes and hydromorphological variations. Therefore, given the range of compositions, sizes, densities, and shapes associated with MPs, reach-scale variations in the riverine systems would determine the distribution of MPs. In this regard, the MPs sampling protocol should be designed to reflect these variations as sampling sediments for MPs from different channel-bed features (morphological units) and associated biotopes (e.g. runs, pools, glides) may yield dramatically different results, even if the samples were collected a few metres apart. As an example, Tibbets et al. (2018) reported that variations in MPs abundance in the River Tame in England were primarily driven by flow velocities (a primary attribute of hydraulic biotopes), with increased particle concentrations reported in areas of decreased flow velocities due to increased deposition of fine sediments and MP. It therefore seemed that MPs concentrations along a river stretch are strongly regulated by the morphological features of the channel and the varieties of hydraulic biotopes present. We therefore propose that MPs sampling protocols should be designed to reflect these hydro-geomorphological considerations.

The second implication is that riverine organisms inhabit specific hydraulic biotopes, and certain species normally prefer specific hydraulic biotopes (Jowett and Richardson 1990; Thomson et al. 2001; Odume 2020). Since the distribution and concentrations of MPs along a river stretch are largely regulated by the prevailing hydraulic biotopes, organismal differential exposure to MPs can be better understood when viewed from this perspective. In this regard, the potential impacts of MPs on different species can be better unravelled through the interaction of MPs physical characteristics and hydraulic biotopes. Future biomonitoring studies should integrate the complex MP-hydraulic interaction for ecologically meaningful investigation into how different organisms could be differentially exposed to MPs, based on their preferred hydraulic biotopes. From a practical perspective, this is useful because following the river classification hierarchy, hydraulic units occur at a practical and ecologically relevant scale in assessing the structural attributes that influence microhabitat character and biotic diversity.

Conclusion

Plastic pollution is a global and pervasive problem in riverine ecosystems. This review highlights the importance of hydraulic biotopes in the distribution and transport of MPs in riverine systems. Hydraulic biotopes are spatially distinct instream flow environments (e.g. run, pool, glide) characterised by specific hydraulic attributes that provide the abiotic environment in which species assemblages or communities live. As the review indicated, the interactions between hydraulic biotopes and MPs physical characteristics are complex, yet critical to understanding the horizontal and vertical transport of MPs, and the potential exposure of instream organisms to MPs. In this regard, we propose explicit hydro-geomorphological considerations in the design of MPs sampling protocols in rivers.

The authors believe that the suggested approach (i.e. the hydro-geomorphological approach using hydraulic biotopes) would pave the way for collecting data that are ecologically meaningful and would simplify our understanding of MPs transport dynamics to facilitate inter-studies comparison. The approach suggested also implies that the diversity of hydraulic biotopes available at any sampling event will vary, with the most remarkable diversity likely to be observed at intermediate discharges.

Declaration of interests

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.