### Analysis of toxic effects and nutrient stress in aquatic ecosystems

Bontje, D.
2010. Analysis of toxic effects and nutrient stress in aquatic ecosystems.

*PHD Thesis* Vrije Universiteit, Amsterdam
#### Abstract

### Chapter 2

We study the effects of toxicants on the
functioning of a phototrophic unicellular organism (an algae) in a
simple aquatic microcosm by applying a parameter-sparse model. The
model allows us to study the interaction between ecological and
toxicological effects. Nutrient stress and toxicant stress, together or
alone, can cause extinction of the algal population. The modelled
algae consume dissolved inorganic nitrogen (DIN) under surplus light
and use it for growth and maintenance. Dead algal biomass is
mineralized by bacterial activity, leading to nutrient recycling. The
ecological model is coupled with a toxicity-module that describes the
dependency of the algal growth and death rate on the toxicant
concentration. Model parameter fitting is performed on experimental
data from Liebig et al. (2008). These experiments were especially
designed to include nutrient limitation, nutrient recycling and
long-term exposure to toxicants. The flagellate species Cryptomonas
sp. was exposed to the herbicide prometryn and insecticide methyl
parathion in semiclosed Erlenmeyers. Given the total limiting amount
of nitrogen in the system, the estimated toxicant concentration at
which a long-term steady population of algae goes extinct will be
derived. We intend to use the results of this study to investigate the
effects of ecological (environmental) and toxicological stresses on
more realistic ecosystem structure and functioning.
### Chapter 3

Since Rosenzweig showed the destabilisation of exploited ecosystems,
the so called Paradox of enrichment, several mechanisms have been
proposed to resolve this paradox. In this paper we will show that a
feeding threshold in the functional response for predators feeding on
a prey population stabilizes the system and that there exists a
minimum threshold value above which the predator-prey system is
unconditionally stable with respect to enrichment. Two models are
analysed, the first being the classical Rosenzweig- MacArthur (RM)
model with an adapted Holling type-II functional response to include a
feeding threshold. This mathematical model can be studied using
analytical tools, which gives insight into the mathematical properties
of the two dimensional ode system and reveals underlying stabilisation
mechanisms. The second model is a mass-balance (MB) model for a
predator-prey-nutrient system with complete recycling of the nutrient
in a closed environment. In this model a feeding threshold is also
taken into account for the predatorprey trophic interaction. Numerical
bifurcation analysis is performed on both models. Analysis results are
compared between models and are discussed in relation to the
analytical analysis of the classical RM model. Experimental data from
the literature of a closed system with ciliates, algae and a limiting
nutrient are used to estimate parameters for the MB model. This
microbial system forms the bottom trophic levels of aquatic ecosystems
and therefore a complete overview of its dynamics is essential for
understanding aquatic ecosystem dynamics.
### Chapter 4

In Chapter 2 we studied the effects of toxicants on the functioning of
phototrophic unicellulars (algae) in a simple aquatic microcosm with a
parameter-sparse model. Now we extend this model to include algivorous
ciliates. The modelled algae consume dissolved inorganic nitrogen
(DIN) under surplus light and use it for growth and maintenance. The
ciliates feed on the algae for growth and maintenance. Dead bacteria,
feeding waste-products and dead ciliates add to a detritus
pool. Detritus is mineralized by bacterial activity, leading to
nutrient recycling. The ecological model is coupled with a
toxicity-module that describes the dependency of each species
biological rates on the toxicant concentration. Model parameter
fitting is performed on experimental data from Liebig et
al. (2008). The flagellate species Cryp- tomonas sp. was exposed to the
herbicide prometryn and insecticide methyl parathion in semi-closed
Erlenmeyers while being preyed upon by either the ciliate Urotricha
furcata or Coleps spetai with the autotrophic endosymbiont Chlorella
sp.. Effects of methyl parathion on Urotricha furcata are directly as
an increased death rate and indirectly via a reduced prey availability
as algal growth was reduced.Coleps spetai with its endosymbiont with
chlorophyll was found to be insensitive to prometryn and only su ered
from food shortage.
### Chapter 5

In this Chapter we study the sublethal effect of
toxicants on the functioning (biomass production, nutrient recycling)
and structure (species composition and complexity) of a simple aquatic
ecosystem in a well-mixed environment (chemostat reactor). The
modelled ecosystem consists of a nutrient consumed by a prey
(e.g. bacteria, alga) which, in turn, is consumed by a predator
(e.g. ciliates, daphnia) population. The dynamic behaviour of this
ecosystem is described by a set of ordinary differential equations
(odes): one for the nutrient and one for each population. The system
is stressed by a toxicant dissolved in the in-flowing water. The
transport of the toxicant is modelled using a mass balance formulation
leading to an ode. Bio-accumulation in the prey and predator
populations is via uptake from the water phase, in case of the
predator also via consumption of contaminated prey. Mathematically
this process is described by a one-compartment model for the kinetics
of the toxicant: uptake (from water and food) and elimination. The
toxicant affects the development of individuals which make up
populations. In the model the physiological parameters depend on the
internal concentration of the toxicant in individuals. Examples of
physiological parameters are cost for growth, assimilation e ciency
and maintenance rate. In this Chapter we use bifurcation theory. In
this way the parameter space is divided into regions with
qualitatively different asymptotic dynamic behaviour of the system. As
logical choice for bifurcation parameters are the strength of the
forcing on the system determined by the input rate of nutrient and
toxicant. Our analysis reveals that the relationship between the
population biomass and the amount of toxicant in the reactor is of
paramount importance. The dynamic behaviour of the stressed ecosystem
can be much more complicated than that of the unstressed system. For
instance the nutrient-prey-contaminant system can show bi-stability
and oscillatory dynamics. Due to the toxic effects a total collapse of
the nutrient-prey-predator-contaminant system can occur after invasion
of a predator, in which case both prey and predator population go
extinct.
### Chapter 6

The dynamical behaviour is analysed of a stressed
(nutrient and toxic stress) aquatic ecosystem (for instance a stretch
of river) consisting of nutrients, biotic pelagic and benthic
communities and detritus pools in the abiotic water body and on the
sediment. We analyse the downstream river reach adjacent to the point
of emission of the toxicant. The toxicant is taken up by the organisms
and can affect different chemical-biotic processes in the populations
thereby changing their biological functioning (assimilation, growth,
maintenance, reproduction, survival). These effects on the behaviour of
the populations induce effects on the ecosystem structure and
functioning e.g. extinction or invasion of species. The dynamic
behaviour of all populations and toxicant concentration in the water,
sediment, detritus and in the organisms that built up the populations,
is mathematically described by a set of ordinary differential equations
(odes), taking intra- and interspecific interactions and transport
from and to adjacent river reaches into account. The set of equations
describing the long-term dynamic behaviour of the ecosystem is
analysed using bifurcation theory to find and continue parameter value
combinations where a population goes extinct or where the ecosystem
shows oscillatory behaviour. As bifurcation parameters we will take
the nutrient input (enrichment) and concentration of the toxicant in
the inflow (toxic exposure). We calculate the boundaries of the range
of the parameters that describe the forcing and loading where the
system behaves similarly, that is the long-term dynamics is
qualitatively the same (equilibrium, oscillatory or chaotic behaviour)
[and areas where the quantities are the same]. In this way we obtain
levels of toxic loading where the abundances of all populations are
the same as in the control case. We will also calculate the boundary
of regimes of toxic stress levels that do not lead to a change in the
composition of the ecosystem and therefore its structure, but where
population abundances and internal toxicant concentrations may change
quantitatively. An integrated approach is used whereby the effects of
toxicological (toxic exposure), ecological (feeding, predation,
competition) and environmental stressors (dilution rate) are analysed
simultaneously. Due to feed-back mechanisms such as indirect effects,
between toxic exposure and physical and ecological processes even low
toxic stress can lead to drastic changes in the ecosystem functioning
and structure.