Modelling Primary Producer Interaction and Composition: an Example of a UK Lowland River

Abstract

Nutrient enrichment and drought conditions are major threats to lowland rivers causing ecosystem degradation and composition changes in plant communities. The controls on primary producer composition in chalk rivers are investigated using a new model and existing data from the River Frome (UK) to explore abiotic and biotic interactions. The growth and interaction of four primary producer functional groups (suspended algae, macrophytes, epiphytes, sediment biofilm) were successfully linked with flow, nutrients (N, P), light and water temperature such that the modelled biomass dynamics of the four groups matched that of the observed. Simulated growth of suspended algae was limited mainly by the residence time of the river rather than in-stream phosphorus concentrations. The simulated growth of the fixed vegetation (macrophytes, epiphytes, sediment biofilm) was overwhelmingly controlled by incoming solar radiation and light attenuation in the water column. Nutrients and grazing have little control when compared to the other physical controls in the simulations. A number of environmental threshold values were identified in the model simulations for the different producer types. The simulation results highlighted the importance of the pelagic–benthic interactions within the River Frome and indicated that process interaction defined the behaviour of the primary producers, rather than a single, dominant driver. The model simulations pose interesting questions to be considered in the next iteration of field- and laboratory-based studies.

Appendix: Model Description

This new primary producer model (PPM) is built on previous models [28, 39] but extends the scope of these models by considering further processes in sufficient detail to represent the key interactions amongst the aquatic primary producer types without becoming the model over complex. The model variables and constants are summarised in Tables 4 and 5, and the model equations are listed in Table 6.

Stream hydrology is modelled following a simple approach used in the INCA models (e.g. [28, 59]). The changes in biomasses of the primary producer types are calculated with mass balance equations in Table 6.

1.1 Main Assumptions and Calculation Methods

Benthic grazers are prominent in short- and medium-retention-time rivers and zooplankton in very long-retention-time rivers. The bulk biomass of algal (invertebrate) grazers is separately simulated to make the model more realistic [60]. The model only simulates a total grazer population assuming that the relative biomasses of available food (i.e. benthic vs. free-floating algae) in the individual reaches will automatically reflect whether benthic and/or free-floating grazers are present. Grazing losses for each primary producer type are calculated by considering their relative biomasses in the total algal biomass (see, for example, Eq. 7 in Table 6). Predation on herbivorous grazers is also included with the approximated presence of fish. The Arrhenius temperature limitation formula is used (Eq. 8 in Table 6) to represent the seasonal abundance of predators of herbivorous grazers (i.e. young fish hatch during spring and summer and they feed increasingly in late summer and autumn, when the water temperature is highest [40]). The grazing of macrophytes is only represented by a simple loss term in the macrophyte mass balance equation (Eq. 18 in Table 6) which includes a term representing presence of predators (i.e. fish eating invertebrates). The model is forced to keep a minimum biomass for both primary producers and grazers to ensure the biomass increment (i.e. re-growth) if conditions are favourable for growth (see, for example, Eqs. 17–19 in Table 6).

Entrainment of organic matter was found to be a key process in the River Kennet during the preliminary data assessment, when at least half of the chlorophyll peaks resulted from turbulent flow conditions and not from true phytoplankton populations [32]. The importance of bioavailable organic matter on food webs is known [61, 41], although some studies found the opposite [62]. The entrained dead organic matter in the current version of the model only affects light limitation, and it does not affect the growth of the herbivorous grazer group. Thus, it was decided that dead organic matter is not modelled explicitly; rather, it is assumed to be an unlimited source for entrainment. Entrainment of dead organic matter is captured in the model with a simple bank erosion formula (Eq. 13 in Table 6) developed by Jarritt and Lawrence [63].

Nutrient limitation is simulated with the Michaelis-Menten formula [64]. It is recognised that benthic and leaf surface boundary layer transfer of nutrients might be important for benthic vegetation in oligotrophic rivers; however, such equations deliberately were excluded for simplicity.

Light limitation is one of the most important environmental factors in the model, which links all different primary producer types (Eqs. 4, 16, 21 and 29 in Table 6). In case of the algae (live suspended algae (SA), epiphytes (E), SB), the Whitehead and Hornberger [39] light limitation formulation is used, which assumes an optimum solar radiation value for each type. The macrophyte light limitation, on the other hand, is simulated with a Michaelis-Menten formula [64], which allows the macrophytes to grow in a wider range of incoming solar radiation (i.e. even in winter). The light attenuation in the water column is illustrated in Fig. 2 in the main text. It is assumed that the stream water is turbulent, and thus the live suspended algae are at the water surface many times each day and therefore do not experience additional light limitation if they spend some time at greater depth. Thus, only the solar radiation level at the water surface and their own self-shading affect their light limitations. At the other end of the spectrum, sediment biofilm is affected by both the macrophyte biomass and the overlying water column. The solar radiation reaching the macrophyte leaves and epiphytes is reduced by the free-floating organic and inorganic particles (SA, EOM and suspended sediment), the epiphyte cover and also by the distance from the water surface. The distance from the water surface to the leaves of the macrophytes varies with water depth, macrophyte biomass and discharge. Table 9 presents the concept of the relative depth estimation within the PPM. Shallow, moderate and deep waters are differentiated based on the reach depth (RD). When the water level is very low (<0.3 m), the light limitation is only increased when the macrophyte biomass is low. Moderate water depths (0.3–2.0 m) are further divided into three categories based on whether base flow (Q < Q ₉₀), normal flow (Q ₉₀ ≤ Q < Q _ave) or storm flow occurs (Q _ave ≤ Q). Finally, when the water depth is greater than 2.0 m, the depth from the surface to the macrophyte leaves is automatically considered as deep. These five criteria describe the flow conditions, but the height of the plant is expressed as a function of the actual biomass and the self-shading coefficient (m _M and k ₂₇, respectively). Self-shading is included in the relative depth calculation, because it represents a threshold for macrophyte abundance, above which severe light limitation occurs. For the same reason, it can represent life stages and hence vertical positions of macrophyte leaves in the water column. Four arbitrary classes are set based on the actual biomass and the self-shading coefficient, representing negligible, small, moderate and large biomasses. This method does not require additional parameters and still provides a dynamic estimate for different conditions. The relative depth values in the table can be adjusted to the characteristics of the dominant macrophyte species of the river. R. penicillatus, for example, can be found commonly with length varying between 1 and 4 m, trailing downstream in the flowing water [52]. It prefers moderate to fast flows and is fully submerged in waters of up to 2 m in depth but concentrated in depths of less than 1 m [65, 66]. Therefore, no light extinction is assumed if the water depth is less than 2 m deep, the discharge is less than the mean annual average and the macrophyte community has high biomasses. The plant is quite flexible, and at higher discharges, in response to the higher drag force, it is forced closer to the substrate; therefore, the effect of light extinction increases. At high discharges, the plant maybe severed at the bottom [67]. All other settings in the table are based on expert judgment.

Natural mortality of macrophytes in the model simulation occurs under three conditions (Eq. 17 in Table 6). The daily death rate (k ₂₄) decreases the macrophyte biomass during low flow conditions. Discharges higher than the reach average increase this natural mortality rate (i.e. Q _out/Q _ave ratio) through breakage. Finally, sediment instability might occur when the discharge of the river increases too quickly from one time step to the next (i.e. Q _out,t /Q _{out,t − 1}) as a result of a storm event. Sediment washout can remove the rooted vegetation [68], especially in the dieback period of macrophytes in August–October [69]. In this model, sediment instability results in the breakage of macrophytes as a further penalization (Mhd) on the natural mortality rate. Sediment instability has an effect on all fixed primary producers (Eqs. 17, 25 and 31 in Table 6). Weed cutting of macrophytes used to be a practice in the UK chalk rivers to help prevent flooding. If the date(s) and extent of weed cutting are known, the PPM model can consider this human influence in the calculations by applying a user-defined percentage macrophyte loss value (H _M) in Eqs. 19 and 27. The loss of epiphytes is assumed to be proportional to the macrophyte losses.

Epiphyte growth is limited by the available surfaces [70] and by self-shading [16]. When the epiphyte layer on the macrophyte becomes too thick, the bottom layer of epiphytes dies (i.e. insufficient light), and the epiphyte layer loses its adhesive properties and will be sloughed off. However, the macrophytes do not benefit from this for long, because a new epiphyte community will colonise these clear areas soon [16]. These process are considered in the space limitation equation (Eq. 21 in Table 6) that compares the macrophyte/epiphyte biomass ratio with a user-defined minimum ratio (Ratio_M:E,min). Below this ratio, the epiphytic growth ceases (no colonisable area). Above this ratio, the limitation is defined by the half-saturation constant for space limitation (k ₃₈).

Sloughing of epiphytes is due to turbulence and increases the biomass of suspended algae in the water column. Zimba and Hopson [71] observed a non-linear change in the removal efficiency rate over time (i.e. quick removal of loosely attached species and difficult removal of more tightly attached species). The Michaelis-Menten formula [64] was adopted in this PPM model to ensure that the tightly attached species are only removed by the most severe currents (Eq. 24 in Table 6). Since turbulence and drag force in the new model are not explicitly simulated, the ratio of the actual discharge and the average discharge is used to characterise the flow conditions in each reach similarly to the sedimentation term (i.e. actual vs. ‘normal’ flow). However, this process is more complicated, because the drag force of the current on the epiphytes quickly decreases if the macrophytes form patches in the river [33] providing shelter for the inner stands and leaves and therefore also for the epiphytic algae. Therefore, the increase in macrophyte biomass decreases the effect of sloughing on epiphytes. This sloughing equation is also important to ensure that no epiphytic growth can occur before the macrophytes start growing.

Based on a review of the literature, there is no clear relationship between the drag force and the sediment biofilm export from the boundary layer [72, 73]. For this reason, the new PPM model assumes that the sloughing of the algal cells of the sediment biofilm is negligible because these cells are located completely in the boundary layer. Space limitation of sediment biofilm is represented with a simple, linear response equation (Eq. 29 in Table 6).

Finally, there is some evidence that the biomass/chlorophyll-a ratio of algae changes with increasing grazing [74], because the algal community shifts to more filamentous species with thicker walls upon severe grazing, regardless of enrichment. To simplify the model, this ratio was assumed to be constant in this model (Ratio_C:Chl-a ).

Table 4 Overview of the ecological parameters (same for all reaches)

Table 5 Overview of the non-ecological parameters and constants

Table 6 Model equations

Table 7 Used model parameters in the Frome application

Table 8 Environmental factors limiting primary production in the River Frome

Table 9 Relative depth approximation of submerged macrophyte leaves from the water surface