Study of the nutrient and plankton dynamics in Lake Tanganyika using a reduced-gravity model

An eco-hydrodynamic (ECOH) model is proposed for Lake Tanganyika to study the plankton productivity. The hydrodynamic sub-model solves the non-linear, reduced-gravity equations in which wind is the dominant forcing. The ecological sub-model for the epilimnion comprises nutrients, primary production, phytoplankton biomass and zooplankton biomass. In the absence of significant terrestrial input of nutrients, the nutrient loss is compensated for by seasonal, wind-driven, turbulent entrainment of nutrient-rich hypolimnion water into the epilimnion, which gives rise to high plankton productivity twice in the year, during the transition between two seasons. Model simulations predict well the seasonal contrasts of the measured physical and ecological parameters. Numerical tests indicate that the half saturation constant for grazing by zooplankton and the fish predation rate on zooplankton affect the zooplankton biomass measurably more than that of phytoplankton biomass. This work has implications for the application of this model to predict the climatological biological productivity of Lake Tanganyika.     

The effects of seston food quality on planktonic food web patterns

In planktonic food webs, the conversion rate of plant material to herbivore biomass is determined by a variety of factors such as seston biochemical/elemental composition, phytoplankton cell morphology, and colony architecture. Despite the overwhelming heterogeneity characterizing the plant–animal interface, plankton population models usually misrepresent the food quality constraints imposed on zooplankton growth. In this study, we reformulate the zooplankton grazing term to include seston food quality effects on zooplankton assimilation efficiency and examine its ramifications on system stability. Using different phytoplankton parameterizations with regards to growth strategies, light requirements, sinking rates, and food quality, we examined the dynamics induced in planktonic systems under varying zooplankton mortality/fish predation, light conditions, nutrient availability, and detritus food quality levels. In general, our analysis suggests that high food quality tends to stabilize the planktonic systems, whereas unforced oscillations (limit cycles) emerge with lower seston food quality. For a given phytoplankton specification and resource availability, the amplitude of the plankton oscillations is primarily modulated from zooplankton mortality and secondarily from the nutritional quality of the alternative food source (i.e., detritus). When the phytoplankton community is parameterized as a cyanobacterium-like species, conditions of high nutrient availability combined with high zooplankton mortality led to phytoplankton biomass accumulation, whereas a diatom-like parameterization resulted in relatively low phytoplankton to zooplankton biomass ratios highlighting the notion that high phytoplankton food quality allows the zooplankton community to sustain relatively high biomass and to suppress phytoplankton biomass to low levels. During nutrient and light enrichment conditions, both phytoplankton and detritus food quality determine the extent of the limit cycle region, whereas high algal food quality increases system resilience by shifting the oscillatory region towards lower light attenuation levels. Detritus food quality seems to regulate the amplitude of the dynamic oscillations following enrichment, when algal food quality is low. These results highlight the profitability of the alternative food sources for the grazer as an important predictor for the dynamic behavior of primary producer–grazer interactions in nature.     

Integrated model projections of climate change impacts on a North American lake

Climate change is likely to impact terrestrial and aquatic ecosystems via numerous physical and biological mechanisms. This study outlines a framework for projecting potential impacts of climate change on lakes using linked environmental models. Impacts of climate drivers on catchment hydrology and thermal balance in Onondaga Lake (New York State) are simulated using mechanistic models HSPF and UFILS4. Outputs from these models are fed into a lake ecosystem model, developed in AQUATOX. Watershed simulations project increases in the magnitude of peak flows and consequent increases in catchment nutrient export as the magnitude of extreme precipitation events increases. This occurs concurrently with a decrease in annual stream discharge as a result of increased evapotranspiration. Simulated lake water temperatures increase by as much as 5 °C during the 2040-2069 time period, accompanied by a prolonging of the duration of summer stratification. Projected changes include shifts in the timing of nutrient cycling between lake sediments and water column. Plankton taxa projected to thrive under climate change include green algae and Bosmina longirostris. Responses for species at higher trophic levels are mixed. Benthic macroinvertebrates may either prosper (zebra mussels) or decline (chironomids), while fish (e.g., gizzard shad) exhibit high seasonal variability without any clear trend.     

Lake of flies, or lake of fish? A trophic model of Lake Malawi

Ecosystem-focused models have, for the first time, become available for the combined demersal and pelagic components of a large tropical lake ecosystem, Lake Malawi. These provide the opportunity to explore continuing controversies over the production efficiencies and ecological functioning of large tropical lakes. In Lake Malawi these models can provide important insight to the effect of fishing on fish composition, and the potential competition that the lakefly Chaoborus edulis may have with fisheries production. A mass-balanced trophic model developed for the demersal fish community of the southern and western areas of Lake Malawi was integrated with an existing trophic model developed for the open-water pelagic. Input parameters for the demersal model were obtained from a survey of fish distributions, fish food consumption studies, and from additional published quantitative and qualitative information on the various biotic components of the community. The model was constructed using the Ecopath approach and software. The graphically presented demersal food web spanned four trophic levels and was based primarily on consumption of detritus, zooplankton and sedimented diatoms. Zooplankton was imported into the system at trophic levels three and four through fish predation on carnivorous and herbivorous copepods and Chaoborus larvae. It is proposed that the primary consumption of copepods was by fish migrating into the pelagic zone. Chaoborus larvae in the demersal were probably consumed near the lakebed as they conducted a daily migration from the pelagic to seek refuge in the sediments. This evidence for strong benthic-pelagic coupling provided the opportunity for linking the demersal model to the existing model for the pelagic community so producing the first model for the complete ecosystem. Energy fluxes through the resulting combined model demonstrated that the primary import of biomass to the demersal system was detritus of pelagic origin (72.1%) and pelagic zooplankton (10.6%). Only 15.8% of the biomass consumed within the demersal system was of demersal origin. Lakefly production is efficiently utilised by the lake fish community, and any attempt to improve fishery production through introduction of a non-native plantivorous fish species would have a negative impact on the stability and productivity of the lake ecosystem.     

Linking species- and ecosystem-level impacts of climate change in lakes with a complex and a minimal model

To study the interaction between species- and ecosystem-level impacts of climate change, we focus on the question of how climate-induced shifts in key species affect the positive feedback loops that lock shallow lakes either in a transparent, macrophyte-dominated state or, alternatively, in a turbid, phytoplankton-dominated state. We hypothesize that climate warming will weaken the resilience of the macrophyte-dominated clear state. For the turbid state, we hypothesize that climate warming and climate-induced eutrophication will increase the dominance of cyanobacteria. Climate change will also affect shallow lakes through a changing hydrology and through climate change-induced eutrophication. We study these phenomena using two models, the full ecosystem model PCLake and a minimal dynamic model of lake phosphorus dynamics. Quantitative predictions with the complex model show that changes in nutrient loading, hydraulic loading and climate warming can all lead to shifts in ecosystem state. The minimal model helped in interpreting the non-linear behaviour of the complex model. The main output parameters of interest for water quality managers are the critical nutrient loading at which the system will switch from clear to turbid and the much lower critical nutrient loading – due to hysteresis – at which the system switches back. Another important output parameter is the chlorophyll-a level in the turbid state. For each of these three output parameters we performed a sensitivity analysis to further understand the dynamics of the complex model PCLake. This analysis showed that our model results are most sensitive to changes in temperature-dependence of cyanobacteria, planktivorous fish and zooplankton. We argue that by combining models at various levels of complexity and looking at multiple aspects of climate changes simultaneously we can develop an integrated view of the potential impact of climate change on freshwater ecosystems.     

Role of the fish Astyanax aeneus (Characidae) as a keystone nutrient recycler in low-nutrient neotropical streams

Nutrient recycling by animals is a potentially important biogeochemical process in both terrestrial and aquatic ecosystems. Stoichiometric traits of individual species may result in some taxa playing disproportionately important roles in the recycling of nutrients relative to their biomass, acting as keystone nutrient recyclers. We examined factors controlling the relative contribution of 12 Neotropical fish species to nutrient recycling in four streams spanning a range of phosphorus (P) levels. In high-P conditions (135 microg/L soluble reactive phosphorus, SRP), most species fed on P-enriched diets and P excretion rates were high across species. In low-P conditions (3 microg/L SRP), aquatic food resources were depleted in P, and species with higher body P content showed low rates of P recycling. However, fishes that were subsidized by terrestrial inputs were decoupled from aquatic P availability and therefore excreted P at disproportionately high rates. One of these species, Astyanax aeneus (Characidae), represented 12% of the total population and 18% of the total biomass of the fish assemblage in our focal low-P study stream but had P excretion rates > 10-fold higher than other abundant fishes. As a result, we estimated that P excretion by A. aeneus accounted for 90% of the P recycled by this fish assemblage and also supplied approximately 90% of the stream P demand in this P-limited ecosystem. Nitrogen excretion rates showed little variation among species, and the contribution of a given species to ecosystem N recycling was largely dependent upon the total biomass of that species. Because of the high variability in P excretion rates among fish species, ecosystem-level P recycling could be particularly sensitive to changes in fish community structure in P-limited systems.     

Evaluation and application of a three-dimensional water quality model in a shallow lake with complex morphometry

Fundamental hydrodynamic and ecological processes of a lake or reservoir could be adequately depicted by one-dimensional (1D) numerical simulation models. Whereas, lakes with significant horizontal water quality and hydrodynamic gradients due to their complex morphometry, inflow or water level fluctuations require a three-dimensional (3D) hydrodynamics and ecological analyses to accurately simulate their temporal and spatial dynamics. In this study, we applied a 3D hydrodynamic model (ELCOM) coupled with an ecological model (CAEDYM) to simulate water quality parameters in three bays of the morphologically complex Lake Minnetonka. A considerable effort was made in setting up the model and a systematic parameterization approach was adopted to estimate the value of parameters based on their published values. Model calibration covered the entire length of the simulation periods from March 29 to October 20, 2000. Sensitivity analysis identified the top parameters with the largest contributions to the sensitivity of model results. The model was next verified with the same setup and parameter values for the period of April 25 to October 10, 2005 against field data. Spatial and temporal dynamics were well simulated and model output results of water temperature (T), dissolved oxygen (DO), total phosphorus (TP) and one group of algae (Cyanobacteria) represented as chlorophyll a (Chla) compared well with an extensive field data in the bays. The results show that the use of the model along with an accurate bathymetry, a systematic calibration and corroboration (verification) process will help to analyze the hydrodynamics and geochemical processes of the morphologically complex Lake Minnetonka. An example of an ecological application of the model for Lake Minnetonka is presented by examining the effect of spatial heterogeneity on coolwater fish habitat analysis in 3D and under a scenario where horizontal spatial heterogeneity was eliminated (1D). Both analyses captured seasonal fish habitat changes and the total seasonal averages differed moderately. However, the 1D analysis did not capture local and short duration variabilities and missed suitable fish habitat variations of as much as 20%. The experiment highlighted the need for a 3D analysis in depicting ecological hot spots such as unsuitable fish habitats in Lake Minnetonka.     

Structure and function of a warm monomictic lake

Construction and simulation of a model of Lake Conway, Florida, U.S.A., provided a framework for defining major characteristics of this ecosystem. The relationships that are formalized in this model comprise a set of hypotheses about the nature of a warm monomictic lake. The data that were used to parameterize the model came primarily from literature estimates, although approximations of biomass levels were available from associated research conducted on the lake.Submersed macrophytes are a major biomass component in Lake Conway; simulation suggested that their role in nutrient recycling overshadows their importance in the grazing food chain. Phytoplankton biomass and degree of fluctuation are considerably lower than are observed in most cool temperate lakes, although simulated respiration and herbivory rates are closer to temperate values than tropical values. Simulated epipelic algae biomass varies an order of magnitude during the year, and this group appears to be a significant part of the food chain. Simulated zooplankton consumption and turnover rates are very high, in part because of the relatively small biomass per individual. Simulation of the model suggests that slightly more carbon is processed through the grazing food chain in Lake Conway than through the detritus food chain.     

Grazing rates of the Atlantic menhaden Brevoortia tyrannus as a function of particle size and concentration

The feeding behavior of adult Atlantic menhaden (Brevoortia tyrannus) upon 5 species of phytoplankton and 2 species of zooplankton has been studied. Four recognizable feeding stages which were a function of the concentration and size of the food particles were observed. During rapid feeding the fish swam at a constant speed for a prolonged period over a wide range of particle concentrations. Particle and food carbon-concentrations at the threshold for initiation and termination of feeding were inversely related to particle size. Carteria chuii (13.2 μ) was not grazed at a significant rate, while two-cell chains of Skeletonema costatum (16. 5 μ) were filtered from the water, indicating a minimum-size threshold for filtration of between 13 and 16 μ. The most rapid filtering rates were observed for the copepod Acartia tonsa ( \(\bar x\) volume swept clear = 24.8 l/fish/min). The maximum food-particle size acceptable to a menhaden appears to be between Acartia tonsa (1200 μ) and adult Artemia salina (10 mm). These results suggest that the large schools of menhaden found in Atlantic coastal waters could have a significant effect on the plankton, selectively grazing zooplankton, larger phytoplankton, and the longer chains of chain-forming diatoms.     

Shift in nutrient and plankton community in eutrophic lake following introduction of a freshwater bivalve

The impact of the freshwater bivalve Corbicula leana on plankton community dynamics was examined during a cyanobacterial bloom period. Nutrient and chlorophyll concentrations, primary productivity, and phytoplankton and zooplankton communities in the experimental enclosures were measured at 2-3 day intervals. The introduction of mussels reduced net primary productivity and phytoplankton and chlorophyll. Chlorophyll decreased immediately following addition of 100 mussels and then increased over time. After 600 mussels were added, chlorophyll decreased continuously from 87to 25 microg l(-1), approaching that in the mussel-free enclosure. Simultaneously, water transparency increased and concentrations of suspended solids and total phosphorus decreased. Mussel addition caused short-term increases in nutrient concentrations, especially following high-density treatment: phytoplankton density decreased, while cell density in the mussel-free enclosure increased. Zooplankton densities in the two enclosures were similar; however, carbon biomass in the mussel enclosure increased, associated with an increase in large zooplankton. The trophic relationship between phytoplankton and zooplankton was positive in the mussel-free enclosure and negative in the mussel-treatment enclosure, possibly reflecting effects of mussels on both consumer and resource control in the plankton community. Thus, filter feeding by Corbicula affects nutrient recycling and plankton community structure in a freshwater ecosystem through direct feeding and competition for food resources.     

Are closure terms appropriate or necessary descriptors of zooplankton loss in nutrient–phytoplankton–zooplankton type models?

Our current knowledge of plankton ecology ascribes a large proportion of zooplankton losses to zooplankton cannibalism and carnivory, rather than via the activity of higher trophic levels beyond the plankton. However, planktonic ecosystem models, such as the widely used nutrient–phytoplankton–zooplankton (NPZ) type models, typically represent all zooplankton losses by mathematically (rather than biologically) justified closure functions. Even where it is assumed that these closure functions include zooplanktonic cannibalism and carnivory, these processes are not explicitly implemented within the grazing function of the zooplankton. Here it is argued that this representation of zooplankton losses through “closure” terms within planktonic food web models is neither appropriate nor necessary. The general consequences of implementing a simple function incorporating zooplankton cannibalism and carnivory (intra-guild predation) within a planktonic food web model are compared against models implementing different types of traditional closure functions. While the modelled biomass outputs may appear similar, the fate of annual primary production and f-ratios vary widely. There appears no justification for the continued use of traditional closure term to depict zooplankton loss processes on biological or modelling arguments. To do so can seriously misrepresent the fate of primary production and thence trophic dynamics.     

Influence of surface slicks on the distribution and onshore movements of small fish

The surface slicks of internal waves were sampled in continental shelf waters off the east coast of Northland, New Zealand, from 1982–1985. Densities of small fish and zooplankton from surface waters were higher in slicks than in the rippled water adjacent to them. Presettlement fish, of species found as adults nearshore, were abundant in ichthyoplankton samples and visual counts from slicks. These fish ranged from fish with primordial fins to those with adult fin-ray counts. Some small reef fish aggregated around drift algae. Drift algae were also abundant in slicks. Slicks moved at 0.5 to 1.25 km h^-1 in the direction of shore. Hence, a consequence of aggregation in slicks is that presettlement fish may be transported onshore. When slicks were absent, drift algae were found in scattered patches at different distances from shore. In the presence of slicks algae were aligned in them. Internal waves, therefore, may have the ability to turn scattered distributions into regular patterns. Because the surface slicks that internal waves produce are common coastal features, we should modify our perspective of how small fish and zooplankton are distributed in local water masses. Furthermore, the occurrence of internal waves should influence the way in which surface waters are sampled. We suggest that fish settlement patterns need to be investigated concurrently with measurements of physical mechanisms which may promote aggregation, and onshore movements of small fish (e.g. internal waves). These waves may contribute to the variability in settlement rates of small fish.     

The impact of detritivorous fishes on a mangrove estuarine system

Sundarban mangrove ecosystem in India is one of the largest detritus based ecosystem of the world and it supplies the detritus and nutrients to the adjacent Hooghly-Matla estuarine complex. In this estuary a group of fish completely detritivorous in nature, belonging to the genus Mugil spp. is present. This group of fish is expected to have important effects on the trophic dynamics of ecosystems, but exact nature of these effects is not known. In order to study the impact that detritivory by fish may have on the estuarine food chain, we developed mathematical formulations. We run two models, one with phytoplankton, zooplankton, carnivorous fish, detritus and nutrient and without this group of fish and a second one after including this fish in the system. In our model this group of fish has no major impact on primary productions of the estuarine system but has extensive role in total fish production. Coexistence of detritivorous fish and carnivorous fish occurs within reasonable parameter range. We have tested different growth rates of phytoplankton, grazing rates and predation rates of zooplankton, detritivorous fish and carnivorous fish for total system equilibrium. Carnivorous fish predation rate on detritivorous fish and detritivorous fish grazing rate on detritus are very important. Different foraging ratios are also tested in this study. Foraging preference of carnivorous fish on detritivorous fish appears significant for the system equilibrium.     

Extrinsic and intrinsic controls of zooplankton diversity in lakes

Pelagic crustacean zooplankton were collected from 336 Norwegian lakes covering a wide range of latitude, altitude, lake area, mean depth, production (as chlorophyll a), and fish community structure. Mean zooplankton species richness during the ice-free season was generally low at high latitudes and altitudes. Further, lower species richness was recorded in western lakes, possibly reflecting constraints on migration and dispersal. However, despite obvious spatial limitations, geographic boundaries were only weak predictors of mean zooplankton richness. Similarly, lake surface area did not contribute positively to mean richness such as seen in other ecosystem surveys. Rather, intrinsic factors such as primary production and fish community (planktivore) structure were identified by regression analysis as the major predictors of zooplankton diversity, while a positive correlation was observed between species richness and total zooplankton biomass. However, in spite of a large number of variables included in this study, the predictive power of multiple regression models was modest (<50% variance explained), pointing to a major role for within-lake properties, as yet unidentified intrinsic forces, stochasticity, or dispersal as constraints on zooplankton diversity in these lakes.     

A three-trophic level estuarine model: Synergism of two mechanistic simulations

Few numerical simulations have attempted to include a high degree of biological detail for several trophic levels. Typically, in planktonic ecosystem models, if the dynamics of nutrients, phytoplankton and herbivorous zooplankton are formulated with ecological complexity, then carnivores are ignored, forced or modeled in an extremely simplified manner. Extensive mechanistic detail for important carnivores is difficult to represent because reliable and relevant ecological data are rarely available for appropriate species and local populations. Further, the wide temporal and spatial differences between life histories of lower plankton and carnivores may be technically difficult to model.In Narragansett Bay, Rhode Island, the ctenophore Mnemiopsis leidyi is an important carnivore to which these objections do not apply. A detailed carbon-based simulation model of this population of ctenophores was developed independently from an ecosystem model of Narragansett Bay which included detailed interactions between phytoplankton, primarily herbivorous zooplankton and nutrients. The interfacing of these two models without changing any of the formulations or values of the coefficients provided a test of the commonly used practice of forcing certain components. Both models were originally constructed with the biomass of a critical compartment forced according to observed data; in the plankton model, ctenophores were forced, and in the ctenophore model, zooplankton were forced.Predicted biomasses for zooplankton and ctenophores in the combined model were similar to the results of the two parent models, but improved relative to the actual field observations. From the findings it appears that the strategy of forcing is valid provided the forced patterns are appropriate and reasonable.     

Fish distributions and nutrient cycling in streams: can fish create biogeochemical hotspots?

Rates of biogeochemical processes often vary widely in space and time, and characterizing this variation is critical for understanding ecosystem functioning. In streams, spatial hotspots of nutrient transformations are generally attributed to physical and microbial processes. Here we examine the potential for heterogeneous distributions of fish to generate hotspots of nutrient recycling. We measured nitrogen (N) and phosphorus (P) excretion rates of 47 species of fish in an N-limited Neotropical stream, and we combined these data with population densities in each of 49 stream channel units to estimate unit- and reach-scale nutrient recycling. Species varied widely in rates of N and P excretion as well as excreted N:P ratios (6-176 molar). At the reach scale, fish excretion could meet >75% of ecosystem demand for dissolved inorganic N and turn over the ambient NH4 pool in <0.3 km. Areal N excretion estimates varied 47-fold among channel units, suggesting that fish distributions could influence local N availability. P excretion rates varied 14-fold among units but were low relative to ambient concentrations. Spatial variation in aggregate nutrient excretion by fish reflected the effects of habitat characteristics (depth, water velocity) on community structure (body size, density, species composition), and the preference of large-bodied species for deep runs was particularly important. We conclude that the spatial distribution of fish could indeed create hotspots of nutrient recycling during the dry season in this species-rich tropical stream. The prevalence of patchy distributions of stream fish and invertebrates suggests that hotspots of consumer nutrient recycling may often occur in stream ecosystems.     

A vertical-compressed three-dimensional ecological model in Lake Taihu, China

A three-dimensional ecological model on the basis of the analyses of environmental characteristics is set up for Lake Taihu, one of the largest shallow lakes in China. The hydrodynamic processes, nutrient cycling, chemical processes and biological processes are integrated in the model. Model state variables include: water current, surface displacement, nutrients of nitrogen and phosphorus, as well as their different forms such as ammonia nitrogen, nitrate nitrogen, phosphate phosphorus, etc., biomasses of macroplankton, phytoplankton, zooplankton and fish, and also the nutrient levels of macroplankton and phytoplankton. A nutrient budget and sediment transformation are also coupled in the model. The data from January 17, 1997 to January 18, 1998 are use to calibrate the model. The model results have shown good agreement with the observations. It implies that the model could be used for the lake environmental management and research for examining the processes and determining the water quality. The reasons of deviations between the modelled results and the observed values are also discussed. There are six factors that explain the deviations of the modelled results from the observed values and they can be grouped into two sets. One set of problems is associated with the standard deviation introduced by sampling and analyses. The second set of problems can be solved by introduction of processes lacking in the present model (resuspension, phytoplankton transportation mode under the wind with low speed, shifts in species composition and varied size of phytoplankton and zooplankton). The latter two processes should be included in the model at a later stage by integration of a structurally dynamic approach into the three-dimensional model.     

Numerical behavior of a zooplankton,phytoplankton and phosphorus system

Several numerical features of a three-dimensional dynamical system which models the three species system made up of phytoplankton, zooplankton and organic phosphorus nutrient in a lake environment are presented. Certain properties of this system, such as the existence of limit cycles, are demonstrated numerically. It is shown that in certain time domains the system may be stiff, thus requiring the use of suitable algorithms.     

Two-way coupling versus one-way forcing of plankton and fish models to predict ecosystem changes in the Benguela

‘End-to-end’ models have been adopted in an attempt to capture more of the processes that influence the ecology of marine ecosystems and to make system wide predictions of the effects of fishing and climate change. Here, we develop an end-to-end model by coupling existing models that describe the dynamics of low (ROMS–N₂P₂Z₂D₂) and high trophic levels (OSMOSE). ROMS–N₂P₂Z₂D₂ is a biogeochemical model representing phytoplankton and zooplankton seasonal dynamics forced by hydrodynamics in the Benguela upwelling ecosystem. OSMOSE is an individual-based model representing the dynamics of several species of fish, linked through opportunistic and size-based trophic interactions. The models are coupled through a two-way size-based predation process. Plankton provides prey for fish, and the effects of predation by fish on the plankton are described by a plankton mortality term that is variable in space and time. Using the end-to-end model, we compare the effects of two-way coupling versus one-way forcing of the fish model with the plankton biomass field. The fish-induced mortality on plankton is temporally variable, in part explained by seasonal changes in fish biomass. Inclusion of two-way feedback affects the seasonal dynamics of plankton groups and usually reduces the amplitude of variation in abundance (top-down effect). Forcing and coupling lead to different predicted food web structures owing to changes in the dominant food chain which is supported by plankton (bottom-up effect). Our comparisons of one-way forcing and two-way coupling show how feedbacks may affect abundance, food web structure and food web function and emphasise the need to critically examine the consequences of different model architectures when seeking to predict the effects of fishing and climate change.