Assessing contaminated agricultural sediment and water quality with the macrophyte Myriophyllum spicatum

(1)

Assessing contaminated agricultural

sediment and water quality with the

macrophyte Myriophyllum spicatum

Marieke Stuijt – 11347880 BSc. Biology Bachelor thesis Supervisor: dr. J. A. Vonk Examiner: dr. M. H. S. Kraak 02-08-2019

1. Abstract

Ecotoxicological risk assessment is indispensable for determining the health of aquatic ecosystems. The health of aquatic ecosystems is mainly based on water quality, since the European Water Framework Directive refers 7 times to sediment over referring to water 373. Sediments are the largest chemical repositories on earth, and can act as sink. A kind of toxicants that could end up in the sediment are pollutants originating from agriculture, such as the herbicide metazachlor. A lot of different organisms are depending on the sediment, such as macrophytes. Macrophytes are the basis of biodiversity in freshwater ecosystems, and herbicide could be toxic to these organisms. Therefore, the aim of this research is to investigate whether agricultural sediment and water hampers the development of the rooting aquatic macrophyte Myriophyllum spicatum, and whether the contamination in the water or sediment affects the macrophyte more. In phase one there is investigated whether contaminated water and sediment from the agricultural site hampers the development of M. spicatum, In the second phase, there is investigated to what extent the herbicide metazachlor in water and in sediment, detrimentally affect M. spicatum. Results showed that the root/shoot ratio of the plants grown in the agricultural sediment was lower than the plants of the reference. In the water bioassay, development of the plants was higher in the water of the agricultural site, compared to the reference. Results were inconclusive to what extent metazachlor is harmful to M. spicatum and whether it is more affected by metazachlor

(2)

found in the sediment or in the water. In conclusion, contaminations in the sediment of the agricultural site in Almere Buiten hampers the development of M. spicatum. However, the water of this location is not limiting the development of this plant. The quality of the water is controlled; however, sediment is not measured whatsoever. This experiment shows that the impact of the sediment is bigger than the impact of the water, so sediments need to be assessed to determine the quality of freshwater ecosystems.

2. Introduction

Toxic risks to freshwater ecosystems are caused by countless unregulated compounds that are present in the environment (De Baat et al., 2018). To determine the health of aquatic systems, ecotoxicological risk assessment is key. The ecological status of ditches in the Netherlands are mainly based on water quality, while a lot of different aspect are impacting the biota of the system which allneed to be assessed (Borja et al., 2004). Sediments are an important component of shallow aquatic ecosystem. However, the European Water Framework Directive only refers explicitly to sediment 7 times over referring to water 373 times (Borja et al., 2004).

Sediments are the largest chemical repositories on earth and sinks for a wide range of chemicals (Adams et al., 1992; Babut et al., 2005). Theycan act as sink as well as a secondary source of toxicants (Brinke, 2015). Sediments that have been exposed to toxicants for a longer period, could become permanently accumulated with the toxicants (Brinke, 2015). This poses a problem, because the sediment can act as a source of pollutants long after the pollution in the water has been abated (Salomons et al., 1987; Knauer et al., 2008). Fluctuations of abiotic conditions, such as pH changes, could lead to

mobilization of the accumulated pollutants. (Salomons et al., 1987). Benthic species might take up these chemicals, which could lead to a pathway for these chemicals, when the benthic species are consumed by higher aquatic life (Adams et al., 1992). However, the impact of contaminated sediment is largely overlooked and understudied, while benthic organisms are depended on the sediment. An example of benthic species are aquatic macrophytes, which provide key functions and thus are the basis of

biodiversity in freshwater ecosystems (Brinke, 2015; Tuníc et al., 2015). They produce organic matter and oxygen, and they provide nourishment, shelter, or substrate for a variety of aquatic organisms (Nuttens et al., 2016). Furthermore, they can modify flow, stabilize sediment or can promote retention of organic matter in water. Still, macrophytes rarely play a role when it comes to regulatory or quality criteria decisions, despite the ecological importance of the macrophytes (Knauer et al., 2008). There is a lack of knowledge regarding the relative importance of sediment exposure for the uptake of toxicants by rooted macrophytes, according to the Aquatic Macrophyte Risk Assessment for Pesticides (AMRAP) (Diepens et al., 2014).

Sediments are complex environmental compartments, due to the diverse structure and composition (Brinke, 2015). This makes it hard to assess the quality. Bioassays provide a realistic

exposure scenario from bioavailable sediment-bound contaminant. In this way, interaction between the sediment and the test organisms are provided (Brinke, 2015). With the outcome of bioassays, the ecological risk of the sampled area can be determined without doing expensive chemical analysis (Vethaak et al., 2017).

Some toxicants are originating from agriculture, due to use of pesticide. Agricultural run-off, spray drift, infiltration or drainage leads to non-negligible contamination of aquatic ecosystems. (Nuttens et al., 2016). Especially with regard to herbicides, it is important to assess how they affect aquatic plants, as herbicides are often toxic to non-target plants too. In the Netherlands, at multiple measuring spots of water management, the norm for water of many pesticides is exceeded (Atlas bestrijdingsmiddelen in Oppervlaktewater, 2019). To illustrate, at 49 measuring locations there were more than 100 exceedances of the dutch “milieukwaliteitsnorm” and ‘maximaal toelaatbaar risiconiveau’ (MKN and MTR, norms used to protect the ecosystem) of the pesticides in 2017 (Atlas bestrijdingsmiddelen in

Oppervlaktewater, 2019). One of the common herbicides found in the water, is metazachlor and is exceeded at 430 measuring points in the Netherlands, based on the JG-MKN and MKN. The MAC-MKN and the JG-MAC-MKN are environmental quality norms, where the JG-MAC-MKN is based on the year average

(4)

and the MAC is based on the maximum acceptable concentration. When the JG-MKN norm is met, organisms can be exposed to this concentration on the long term, while the MAC-MKN protects against short term concentration peaks (Vonk, 2013). In 2015, the JG-MKN norm is significantly lowered in the Netherlands (from 34 to 0.8 μg/L) (Roex et al., 2016), which resulted in more exceeding of the norm. As a result, metazachlor was ranked number 1 pesticide in 2015 found near cultivations, based on the amount of exceedance of the MAC-MKN (Roex et al., 2016). Since metazachlor is hydrophobic, it could also be accumulated in the sediment (Diepens, 2014). Still, there is no guideline of safe concentration that the sediment could contain, and the effect of those compounds in the sediment on the living biota, especially the benthic species living at the sediment-water interface remains unclear.

Therefore, it is important to investigate what the effect is of pesticides on the aquatic

macrophytes in aquatic systems, and especially what the effect of contaminated sediment is. The aim of this research is to investigate to what extent contaminated agricultural sediment and water hampers the development of rooting aquatic macrophytes. The research question for the first phase is; does the contaminated water and sediment from the agricultural site hamper the development of the

macrophyte Myriophyllum spicatum, and does the contamination in the water or sediment affect the macrophyte more? For the second phase, the research question is: to what extent does the herbicide metazachlor in water and in sediment, detrimentally affect Myriophyllum spicatum? To this purpose bioassays are performed with Myriophyllum spicatum, where they are subjected to contaminated water and sediment from an agricultural site after which the toxicity of the herbicide metazachlor to the plants is assessed.

3. Materials & methods

3.1 Outline of the study

This research consisted of two phases. In the first phase, M. spicatum was grown on agricultural sediment and in water from the agricultural site. The sediment-free Myriophyllum spicatum bioassay and the water-sediment M. spicatum bioassay are performed and compared with each other (OECD, 2014a; OECD, 2014b).

In phase two of this experiment, there was examined using a toxicity test to what extent herbicide metazachlor limits the development of M. spicatum (OECD, 2014b). This was investigated by growing M. spicatum with different concentrations metazachlor in the water and in the sediment. All the experiments took place in the same growth chamber and are run for 28 days to allow for differences in development of the plants.

3.2 Sampling sites and environmental sampling

Kottertocht is the site that the sediment and water samples were taken from on 15-04-2019 (Appendix a). Accessory coordinates of this site are 52°24'49.3"N 5°15'45.8"E. This is a ditch in Almere Buiten, which is surrounded with multiple greenhouses and an example of an area with potentially a lot of agricultural run-off. This is nearby a measuring point of Dutch water managers, named Kottertocht (Atlas Bestrijdingsmiddelen in Oppervlaktewater, 2019). At this point, the sum of total norm based on the JG-MKN of exceeding of multiple pesticides was more than 100, in the 2016. At this location, metazachlor was found in the year of 2016 and 2017 (Atlas Bestrijdingsmiddelen in Oppervlaktewater, 2019). As

(5)

reference, sediment and water were taken on 17-04-2019 from a ditch in Amsterdam, near Science Park. This site is regarded as clean from pollutants.

To obtain the sediment, cores were collected with a gravity corer at the agricultural site. At the reference site, cores were taken by hand. The sediment that was obtained, was taken out of the core using a core cutter, and the upper ten centimeters of the core was put through a sieve with a mesh size of 0.5 mm and collected in a bucket, after which it was merged. Buckets were filled with water at the location to collect water samples. When the water samples were used for the experiment, the water was mixed to redistribute the suspended matter. Before the collecting of the samples, oxygen concentration and saturation, temperature and the conductivity of the water were measured, using the HQ440D Laboratory Multi-Parameter Meter.

3.3 Test species and culture conditions

The rooting aquatic plant that was used in this research is M. spicatum. This macrophyte was chosen because there are standardized tests available for this species, a sediment-bound toxicity test and a water-bound toxicity test (OECD, 2014a; OECD, 2014b).

The M. spicatum. plants used in phase 1 in both experiments and phase 2 for the sediment test were grown in the lab under the same conditions and the plants used for these experiments were the approximately the same age and size at the beginning of the experiment. The original plants were grown in the lab from August 2018. The tests are conducted with healthy shoot apices of these culture plants that do not have any side shoots, and the size of each shoot apices that is used for the

experiment, is 6 cm. Due to lack of plants, a new set of M. spicatum plants were used for the water test of phase 2. Those were not cultivated in the lab, but provided by a commercial supplier, via the web shop schonevijver.nl (Schonevijver.nl, 2019). Still, the size of the plants was the same compared to the other test in the beginning of the water test.

The tests were set up in a way that they meet the requirement of the OECD (OECD, 2014a; OECD, 2014b), with slight modifications. Because sediment from an agricultural site was tested, it was better to take samples of sediment from nature to use as reference, in order to keep the test conditions as similar as possible. Therefore, there was also no artificial sediment used, to ensure all the sediments contained natural nutrients and microbial communities.

Another modification of the protocol of the OECD is the medium that was used. Dutch standard water (DSW) was used as medium, according to the annex of Wester & Jos (2003), with some slight deviations (Appendix b), to exclude that the growth limitation is caused by contamination of the water. The plants in all tests were grown in a growth chamber, of which the temperature was approximately 20 degree Celsius, and the light regime consisted of 16 hours day and 8 hours night. All replicates of M. spicatum participated to the experiment, and none were used to establish the root growth.

The abiotic conditions of every test were measured every week, starting on day 0 of the tests, to see whether the test conditions were still the same. Therefore, the abiotic measurements consisted of measurement of the pH, temperature, dissolved oxygen concentration and saturation, conductivity and voltage using the HQ440D Laboratory Multi-Parameter Meter. Allowed deviations of the temperature were two degrees Celsius. The pH was not allowed to increase by more than 1.5 units. However, when other validity criteria are met, the test still could be considered as valid when this happens (OECD 2014a, OECD 2014b).

(6)

3.4 Phase 1: Bioassays

During the sediment test, M. spicatum was grown on sediment from the agricultural site. The water test contained water from the agricultural site and no sediment. Both tests consisted of 14 replicates, and each of them was grown in their own cores, that were put in glass aquaria (Figure 1). For the bioassay, plastic cores were used since no hydrophobic substances had to be taken into account.

3.4.1 Bioassay: sediment test

The ability for M. spicatum to grow on the sediment that is obtained in Almere Buiten, close to the agricultural land, was tested with the sediment-bound test. Besides, it is tested if this sediment limits the growth development of the plant more than the water of this area.

After the agricultural sediment samples were obtained from the gravity corer, ten centimeters of the sediment core was cut using a core cutter, and this ten centimeters tall sediment core was

transferred into a smaller core of fifteen centimeters, with a surface of approximately 28 cm2_{, which was}

ultimately used in the experiment. In this way, the agricultural core were kept intact, so that different layers of the sediment were not mixed to present the natural environment as good as possible. The reference sediment was already merged and in bucket. To obtain these sediment cores, the plastic cores of fifteen centimeters, were filled with this sediment till ten centimeters. Once the cores were

transferred, six centimeters shoot apices of M. spicatum were cut from the culture and planted circa 2.5 centimeters deep in the sediment of all cores. During the planting, dehydration of the shoot apex is prevented by keeping it moist. When the apex was planted, the core was filled with DSW, after which the cores were put into an aquarium. Both locations got their own aquarium. The cores in the aquaria are filled with DSW, till two centimeters from the topside of the aquarium.

3.4.2 Bioassay: water test

In what way the water of the agricultural site influences the development of the plant was tested with the sediment-free M. spicatum bioassay test. The cores were filled with the water of the accessory location, till the cores were full. The cores were put in aquaria, but the aquaria were not filled with medium. After the abiotic conditions of the water test were measured, water is refreshed, by disposing the current water and filling the cores with remaining water from the buckets that were filled during the sample collecting, two times a week, starting on day 4, with the second refreshment on day 7. The water that was used to refresh, was collected on the sampling day, except for the last two refreshment of the reference, were water is used that is obtained on 13-05-2018.

(7)

Figure 1: The bioassay. a) the water test b) the sediment test. Of each of the cores are 14 replicates of each location.

3.5 Phase 2: Toxicity tests

Phase two is executed to find out what the effect of herbicide metazachlor is and at which concentration it harmful is to M. spicatum Toxicity tests with sediment and water are performed according to the OECD guidelines (OECD 2014a, OECD 2014b). The norm based on the MAC-MKN, is 0.48 μg/L (Vonk et al., 2013). According to the RIVM, the EC50 of Myriophyllum heterophyllum to metazachlor is 80 μg/L (Vonk et al., 2013). Based on this number, the following concentrations are chosen: the norm, 5 times the norm, 20 times the norm, and 200 times the norm. It is expected that no effect will occur when M. spicatum is exposed to the norm, while there is expected that there might be an effect when exposed to 200 times the norm, compared to the reference condition. Each concentration has 5 replicates, for both the sediment test and the water test (Figure 2).

The reference used for this experiment is the same reference as the sediment test of phase one. So the toxicity tests consisted of four concentration of spiked sediment or water, plus the reference. Since the reference was already run during the bioassay, only the four spiked concentrations of the tests were treated at the same time. Science park water does not contain metazachlor (M. L. de Baat, personal communication, 11 June 2019).

3.5.1 Spiked sediment test

Science park sediment is spiked with metazachlor and a sediment test is run according to the OECD guidelines (OECD, 2014a), with slight modifications. It is desired that the amount of metazachlor that is added to the sediment will result in sediment that has taken up the same amount of metazachlor as when it is in equilibrium with water that consist of one, five, 20 and 200 times the MAC-MKN norm (Vonk et al., 2013). Calculation have been made on what amount will attach to the sediment, based on a Koc of 109.64 (log Koc of 2.04), which is a median of 25 values derived by Vonk et al.(2013). The

(8)

metazachlor is taken up by the sediment and the sediment is dried again to let the methanol evaporate before the sediment was used. After this, the sediment was put into flasks, and filled with approximately 500 mL water and put on roller bank for four days to mix the sediment, so the metazachlor would be distributed over all the sediment.

To determine the concentration of metazachlor after spiking of the sediment, a silicone polymer sheet was employed as passive sampler. Passive sampling is a way of toxicity profiling. Hereby, it is possible to assess potential harmful effect of toxicants that are not included in monitoring programs, identify sources and to investigate relationships between toxicants and ecological effects (Vethaak et al., 2017). The silicone polymer sheet was added to the flask in which the spiked sediment was mixed, before it was put onto the roller bank. It took up the spiked pore water. After four days, the silicone polymer sheet was taken out of the flask and briefly rinsed with demi water. Of this sheet, a piece of one gram was cut of and the metazachlor was extracted from this piece using acetonitrile. The acetonitrile extract was diluted two times with pure water. The metazachlor concentration of the obtained dilution is detected with liquid chromatography–mass spectrometry (LC-MS). With this concentration, the pore water concentration was derived using the octanol/water partition coefficient (log Kow) of 2.49 (Australian Pesticides and Veterinary Medicines Authority, 2016) (PPDB, 2018). With the pore water concentration there is calculated what amount of metazachlor was located in the water, to derive what eventually was taken up by the sediment.

The spiking resulted into four types of sediment with its own concentration of metazachlor. For each concentration, a rectangular aquarium is filled with 5 replicates. Each aquarium contained

approximately 2.4 kg spiked sediment, with a maximum deviation of 166 gram. The surface of the aquaria is approximately 500 cm2_._{Due to a shortage of rectangular aquaria, the sediment which contains}

a concentration of 200 times the norm, was divided between two aquaria, which have a surface of 170 and 310 cm2_{, respectively. To give each replicate the same room to grow, 2 replicates are put into the}

aquarium with a surface of 170 cm2_{, and 3 replicates are put into the aquarium with a surface of 310}

cm2_{. The height of the sediment layer was in all aquaria approximately 5 centimeters, although the} height of the sediment layer in the aquaria of 200 times the norm was slightly more, because of the smaller surfaces.

3.5.2 Spiked water test

For this test, a different culture of M. spicatum is used. These macrophytes were not cultured in the lab and were already flowering. The reference has 5 replicates and consist of non-spiked Science Park water. The plants are grown in glass jars, to prevent that metazachlor attaches to the wall. The volume of these jars is 353 ml. Flask were filled with 2L for spiking. To fill flasks with the water for the spiking, the upper layer of the Science Park water samples was separated and then mixed. In this way, most of organic matter remained in the original samples. Water was spiked by adding the right amount of metazachlor to these flasks. After this, the flasks were put on a roller bank for at least 24 hours. The shoot tips that were cut off from the culture, were put in glass jars, one per jar, and the spiked water was added to until the beginning of the convex of the jar. The spiked water was refreshed once a week. Before the refreshment, abiotic conditions were measured. The water that was used to refresh, was collected on the sampling day, except for the last two refreshment. For the second last refreshment, the water was obtained on 13-05 and for the last refreshment, the water was obtained on 20-13-05.

(9)

Figure 2: The toxicity test. a) the spiked water test b) the spiked sediment test. Of each of the concentrations are five replicates.

3.6 Endpoints

After running the experiments for 28 days, the plants were harvested from the water tests (both the water bioassay as the spiked water test) by pouring the water with the plant out of the cores and the jars. This was carefully done in case any parts of the plants were detached. The plants were harvested from the sediment bioassay by emptying the cores in a sieve, after which the sediment was carefully removed from the sieve, until the plant was exposed and taken out of the sieve. Afterward the sieve was checked on any plant parts that were detached during the harvesting, and the plants were saved in a jar with water at room temperature. The plants are put in a petri dish with water at room temperature to remove the residentially contaminations using a tweezer.

Limitations of development were defined as little or no shoot growth, less root development, less side branches and less fresh and/or dry weight in comparison with the control group. The only variable that differed between the plants used for the experiment prior to the test was the weight of the replicates. Therefore, there is chosen to measure fresh weight and estimate the dry weight also before the experiment is run, to estimate the development of the plant. After harvesting the plants endpoints were measured. Weight is a good indicator of how much a plant has grown. By making use of the dry weight, the measured weight is not dependent on how much water is held by the plant, resulting dry weight to be a more precise measurement. However, it is not possible to obtain these data prior to the experiment. Therefore, the decision has been made to measure both fresh weight and dry weight. To estimate the dry weight at the beginning of the experiments, the dry weight of plants of the same age

(10)

and length as the plants used in the experiments of phase one and phase two the sediment test, was measured. These data were used to calibrate the dry weight of the plants in the beginning used in the experiments. The final dry weight is obtained after drying the plants in a stove for at least 2 days at 70 degree Celsius. The initial weight for both fresh and dry weight are subtracted from the final weight to gain the increase in weight. Side side shoots are counted at the end of the experiment. Side shoot were defined as axillary buds, apical buds, small side shoots or full-grown side shoots, and all were counted. ImageJ is used to measure the total root and total shoot length of the plants (Rueden et al., 2017; Appendix c). The total root length is divided by the total shoot length to obtain the root/shoot ratio. According to Knauer et al. (2008), the endpoints gained fresh weight, gained dry weight and total root length of M. spicatum grown are stable under standard condition and variance between those endpoints are not caused by chance. Therefore, minor effects of pollutants can be detected.

The visuals of the plants and the condition of the test are reported during the abiotic

measurements, in order to observe the process of the tests. Observed aspects are the color of the plant, the firmness of the plant and the color of the water. Any particularities, such as necrosis of chlorosis that is observed is noted, as well as the presence of any other organisms, such as algae, daphnia or worms.

3.7 Data analysis

The results of the bioassays will give more insight about the impact of the contaminated sediment and the contaminated water of the agricultural site. Comparison of the difference between the reference and the agricultural replicates of the sediment and the difference between the reference and the agricultural replicates of the water test will show whether the water or the sediment of the agricultural site limits the development of M. spicatum (more). Statistical analysis is done with R (R Core Team, 2018).

Significance between the different locations is tested with a two sample t-test for the bioassay, provided that the assumptions were not violated. In the case that normality is violated, a Whitney Mann U test is executed. In the case that variances were not equal, Welch t-test is used. To investigate whether metazachlor in water and in sediment affect M. spicatum, the plants of the spiked treatments are compared with the accessory references. The reference and different concentration spiked sediment are tested with the use of an ANOVA, provided that the assumptions are not violated. The post hoc test that is used is the Tuckey’s Honestly Significant Difference (HSD) test, to investigate which groups differ significantly from each other.

4. Results

The results of the measurements of the abiotic conditions showed no extreme fluctuations, and the required conditions according to OECD (2014a, 2014b) were met. The growth conditions are considered as stable (Appendix d). Another requirement was that the plant should double its length during the first 14 days after exposure (OECD, 2014a; OECD, 2014b). This was not the case with the spiked water test. The other tests met the requirements and were considered as valid.

4. 1 Visual findings

Sediment bioassay

At the start of the experiment, the sediment of plant 1 and 7 of the agricultural location contained some hard organic material. When analyzing the plants after one week, it appeared that plant 10 of the

(11)

agricultural condition had not been put in the sediment correctly. Plant 6 was also detached, but still had roots that were connected to the sediment. Plant 3 of the agricultural condition turned out to be not a shoot tip but a shoot that was already cut off before. The same applies for plant. A lot of worm were present, crawling out of the agricultural sediment, during the whole experiment. Other organisms that have been spotted in the agricultural condition are minuscule lobsters (Appendix e, Figure 10). After 14 days, a green layer has formed (probably caused by algae) on the surface of the water. This layer was also observed at the aquarium of the reference sediment test and disappeared after a week, while the layer on the surface of the agricultural sediment test remained there (Appendix e, Figure 11). During

harvesting of the plants, there is presumed that the organisms that were spotted were worms, minuscule lobsters, Asellus aquatics and there might have been a Daphnia. These organisms were not furtherly identified. The part of the plant that was in the sediment came out black. This was accurate for all the plants of both sediment tests.

Water bioassay

When analyzing the plants after one week, algae grew on the plugs of the cores of the reference

(Appendix e, Figure 10). Algae also grew on the plugs of the agricultural location, but in smaller amount. When the plants were harvested, in the core of plant 2 of the agricultural site there was presumed that the organism that were spotted were Daphnia and there might be another kind of Cladocera and snails, that probably originated from the plant culture that was cultivated together with snails. Algae layers were detected on the side of the cores of the plants 3, 13 and 14 of the agricultural site (Appendix e, Figure 11).

Spiked sediment

The shoot tip of 20 times the norm 5 was cut off before or planted in the sediment upside down. On day 27 of the spiked sediment test, there was discovered that the water level had dropped. Consequently, the tips of the plants of the norm 2 and of 5 times the norm 3 and 5 were scorched. During the harvesting of the plants, some root parts remained in the sediment. It was not possible to track down the origin of these roots. After the plants had dried in the stove, something that looked like sand was discovered on top of the plants. It is assumed that this were chemicals that were clumped together.

4.2 Bioassay with sediment

The M. spicatum endpoints gained fresh weight, gained dry weight, side shoots, total shoot length, total root length and root/shoot ratio were compared between the reference sediment and the agricultural sediment (Figure 3). The T-test compared the replicates of the references with the agricultural site for the sediment bioassay (Table 1). Although not significant, the means of the endpoint of the reference were higher than the means of the agricultural condition. However, the difference was only significant when it comes to the root/shoot ratio of the two conditions. The results of the bioassay with sediment shows that the root/shoot ratio of the agricultural site is lower than the root/shoot ratio of the

reference.

Due to a mistake, the shoot length of the replicates 13 and 14 was lost. This result in 12 replicates for the reference for the endpoints total shoot length and root/shoot ratio.

(12)

Figure 3. Boxplots of the bioassay with sediment. Comparison between the reference and the

agricultural site of a) gained fresh weight, b) gained dry weight, c) side shoots, d) total shoot length, e) total root lengths and f) root/shoot ratio. N=14, expect for the endpoint total shoot length and

root/shoot ratio of the reference, here N=12.

a) b) d) c) e) f) (mg) _(mg) (cm) (cm)

(13)

4.3 Bioassay with water

The T-test compared the replicates of the references with the agricultural site for the water bioassay (Table 1). The results of the reference sediment and the agricultural water of gained fresh weight, gained dry weight, side shoots, total shoot length, total root length and root/shoot ratio are displayed with boxplots (Figure 3). These show that the mean of the agricultural replicates is higher for all the variables except for the side shoots. The difference between the two conditions was significant when it comes to the total shoot length, total root length and the root/shoot ratio. Thus, the plants developed better in the water of the agricultural site than in the water of the reference.

Figure 4. Boxplots of the bioassay with water. Comparison between the reference and the agricultural site of a) gained fresh weight, b) gained dry weight, c) side shoots, d) total shoot length, e) total root lengths and f) root/shoot ratio. N=14

Klik om stijl te

bewerken

(mg) _(mg) (cm) (cm) a) _b) c) d) e) f)

(14)

Table 1. Statistical differences (T-tests) for the used M. spicatum endpoints between reference and agricultural site. P-values are shown. N=14*

Sediment bioassay Water bioassay

Gained fresh weight 0.410 0.761***

Gained dry weight 0.330 0.445

Side shoots 0.164 0.812**

Total shoot length 0.632 5.164e-03

Total root length 0.197 0.010

Root/shoot ratio 0.036 4.532e-05**

= Significant difference

* Expect for the endpoint total shoot length and root/shoot ratio of the reference of the bioassay, here N=12.

** Normality was violated, Mann Whitney U test was used. *** Variances not equal, Welch t-test is used.

4.4 Spiked sediment

The water concentration was calculated with the log Kow and the derived concentration of metazachlor on the silicone polymer sheets, that was obtained according to the results of the liquid chromatography– mass spectrometry (Table 2). When the amount of metazachlor on the passive sampler and in the water was known, the concentration in the sediment was calculated (Table 2). The desired concentrations were not exactly met, however, the actual concentrations were a good representation of the intended

concentrations. Contrary to the difference in the pore water concentration, where the desired concentration are not reached.

Table 2. Concentrations of metazachlor in the passive sampler, the water and the sediment. Desired concentration in pore water (μg/L) Concentration in pore water (μg/L) Desired concentration in sediment (μg/kg) Actual concentration in sediment (μg/kg) Factor of increasement with lowest concentration Norm 0.48 0.09 1.64 1.68 1.00 5x Norm 2.4 0.35 8.20 7.82 4.65 20x Norm 9.6 1.85 32.79 30.78 18.32 200x Norm 96 7.07 327.90 294.64 175.38

The results of the one sampled ANOVA’s show the difference of the six variables between the different groups. The groups which show the most significant difference are all the spiked groups, except the 200x norm with the reference, together with the results of the comparison of the 5x norm and 200x norm (Table 3). Most graphs shows a decrease from a concentration of 0 (reference) to 7.82 μg/kg (5 times

(15)

the norm), while an increase of the variable is shown when the concentration increases (from 5 times the norm, to 200 times the norm; 7.82 μg/kg to 294.62 μg/kg) (Figure 4). There is no trend observed, since the condition 5 times the norm differs the most from the reference, then 20 times the norm, then the norm. 200 times the norm differs only with the variable root/shoot ratio. The progression of the mean of the variables over the different concentration shows that there is no dose- effect relation (Figure 5).

Klik om stijl te

bewerken

(mg) (mg) a) b) c)

(16)

Figure 5. Boxplots of the toxicity test with spiked sediment. Comparison between the reference and the agricultural site of a) gained fresh weight, b) gained dry weight, c) side shoots, d) total shoot length, e) total root lengths and f) root/shoot ratio. N=14 for the reference. N=5 for the other treatments.

(cm)

d)

e)

(17)

500 1000 1500 2000 0 5 10 15 sqrt [Metazachlor] (μg/kg) G ai ne d fr es h w ei gh t ( m g) 0 20 40 60 80 0 5 10 15 sqrt [Metazachlor] (μg/kg) G a in e d d ry w e ig h t ( m g ) 17 3 6 9 0 5 10 15 sqrt [Metazachlor] (μg/kg) S id e sh oo ts a) 1-7-2019 •

Klikken om de ondertitelstijl

van het model te bewerken

b)

(18)

0 25 50 75 100 125 0 5 10 15 sqrt [Metazachlor] (μg/kg) T ot al ro o t l en gt h (c m )

Figure 6. The progression of the mean of the variables a) gained fresh weight, b) gained dry weight, c) side shoots, d) total shoot length, e) total root lengths and f) root/shoot ratio fthe square root

18 1 2 0 5 10 15 sqrt [Metazachlor] (μg/kg) R oo t/s ho o t r at io 10 20 30 40 50 0 5 10 15 sqrt [Metazachlor] (μg/kg) T ot al s ho o t l en gt h (c m ) d)

(19)

concentration of metazachlor. Error bars represent standard deviation. N=14 for the reference. N=5 for the other treatments.

Table 3. Results of the ANOVA followed by the Tukey’s HSD test. P-values are shown. Gained

fresh weight

Gained dry weight

Side shoots Total shoot length Total root length Root/shoot ratio Norm- Reference

0.012 180e-03 0.361 0.260 114e-03 586e-04

5x Norm – Reference

473e-05 537e-05 341e-03 0.017 122e-04 101e-04

20x Norm – Reference

842e-04 102e-04 0.146 475e-03 891e-05 707e-05

200x Norm – Reference 0.824 0.336 0.663 0.898 0.163 228e-03 5x Norm – Norm 0.454 0.823 0.414 0.804 0.960 0.980 20x Norm – Norm 0.921 0.905 0.992 0.547 0.937 0.960 200x Norm – Norm 0.279 0.329 0.088 0.121 0.442 0.992 20x Norm – 5x Norm 0.906 1.000 0.680 0.992 1.000 1.000 200x Norm – 5x Norm 704e-03 0.042 116e-03 0.010 0.147 0.854 200x Norm – 20x Norm 0.057 0.064 0.033 340e-03 0.123 0.799 = Significant difference 4.5 Spiked water

The spiked water test failed (Appendix g). This is due to the use of M. spicatum that was not cultivated in the lab, and from a different time. This culture was already flowering, which should not be used

according to the OECD (OECD, 2014b). To compare different condition the macrophytes have to be the same age (OECD, 2014b). Another reason why this test is considered as failed is because the duplication of the mean total shoot length and mean total shoot fresh weight that should happen (OECD, 2014b), failed to appear in the control. Lots of different organisms grew in the water, and the M. spicatum was in some cases surrounded with algae, which also is at the expense of the validity of the test (OECD, 2014b; Appendix g).

(20)

5. Discussion

5.1 Agricultural impact of sediment and water on M. spicatum

The effect on M. spicatum of the agricultural site compared to the reference is caused by the sediment, and not by water. Agricultural water only causes positive growth. It is expected that faster plant

development was due to a higher nutrient availability in the agricultural water than the reference water, especially considering M. spicatum appears mostly in eutrophic water (OECD, 2014b). That the plants were evolving well, does not mean the water is clean. Although it is a good indicator that there were no lack of nutrients and possible toxic compounds did not hamper the plant’s development.

Since the root/shoot ratio is significantly lower in the plants that grew in agricultural sediment, the development of the roots is more hampered than the development of the shoots. According to McCann et al. (2000) roots are the most sensitive of all endpoints. Presumably, the harmful chemicals are more located in the sediment than in the water. The toxicants ended up in the water, and though the water accumulated at the bottom. The roots take up more toxicants from the sediment through the roots than from the water through the shoots. In this way, the roots are the most exposed to the toxicants, and the rest of the plant is exposed to these toxins when the roots take up organic compounds from the

sediment, to which toxicants are bound to (Diepens et al., 2014).

5.2 The impact of metazachlor in sediment and water on M. spicatum

The results of the spiked sediment test were remarkable, since in general, the plants performed better on the spiked sediment with higher concentrations (Figure 4). There is assumed that this might be caused by failing to spike the sediment or to the different growth conditions of the plants.

It is not known how metazachlor is located and how it behaves when it comes to water-

sediment equilibrium. So there is a chance that the spiking of sediment did not result in exposure to the plants of the intended metazachlor concentrations. However, based on the result of the liquid

chromatography–mass spectrometry, it might be assumed that the spiking of the sediment was successful, because it is quite certain that the desired concentrations are adsorbed to the sediment. However, it was expected that the desired concentration in the sediment would match the norm, 5x norm, 20x norm and 200x norm in the water. That this is not the case, can be caused by multiple possible reasons. Prior to the experiment, the Koc was used as an indication to estimate the partition of

metazachlor in water and sediment. The concentrations that were desired in the sediment were based on the Koc. There are resources that provide different values of the Koc. For example, the Pesticide Properties DataBase gives a value of 54.0 for the Koc, which differs than the values used in this

experiment, namely 109.64 (PPDB, 2018) (Vonk et al., 2013). So if the Koc is not representative for the real partition of metazachlor in water and sediment, the aim for the intended concentration was not correct. After the experiment, the amount of metazachlor in the water was calculated based on the concentration that the silicone polymer sheet contained and the Kow. The results of the estimated water concentration show that the pore water concentration is relatively low. For instance, the condition 20x norm contained only approximately four times the MAC-MKN norm, and 200 times the norm contained only 15 times the norm (Table 2). Multiple sources provided the logKow of metazachlor of 2.49 (PPDB, 2018) (Vonk et al., 2013). However, according to Ahrens et al., (2015) the logKow is not exactly the same

(21)

as the partition coefficient for the silicone rubber with water (Ksr-w). They provide a Ksr-w of 3.0. Based on this number, the concentration of metazachlor in water is even lower. So by the use of Koc and Kow, the spiking is based on rough estimation. Moreover, the Koc and Kow can differ between different kind of soils (Ahrens et al., 2015; Allen & Walter, 1987). The concentration herbicide in pore water is a much lower concentration, than the intended concentration, while this concentration is most relevant for the exposure to the plants (S. Droge, personal communication, 11 June 2019). To conclude, the values of the Koc and/or the Kow were not precise enough to estimate the amount of metazachlor that should be added or to estimate the concentration of metazachlor in the water. This resulted in a lower

concentration metazachlor in pore water than intended.

Since metazachlor is relatively hydrophobic, thus a sediment-bound chemical, multiple exposure routes are available(Diepens, 2014). Transport trough the phloem and xylem depends on solubility and hydrophobicity (Diepens, 2014). It is expected that this causes a high translocation potential by dissolved organic carbon and a less translocation potential by water. Dependent on the exposure routes, a

herbicide in the sediment might be more harmful when it is stored in the pore water than bound to the sediment itself. The metazachlor in the sediment could be exchange with the water since the sediment can act as a source of toxicants (Brinke, 2015). Therefore, there was a toxicity test performed for both water and sediment. The results of the spiked water test could have been useful to compare those with the results of spiked sediment test. It could have been noticed how the metazachlor was located and distributed. In the spiked water test, there is no distribution of metazachlor in a different medium, so in this test, it would be clearer to what amount of metazachlor the plants have been exposed, and what the effect of this concentration in the water would be to the development of the plants. Now, the

metazachlor in the toxicity test with sediment is divided over the water and sediment under a ratio that can only be estimated. So it is not clear whether only the metazachlor in the sediment has an impact on the plants or also the metazachlor in the water.

The results do not give any certainty to which norm the plants are exposed. However, the different concentrations are a series at which the concentration is increased, with a factor of circa 5, 20 and 200. There is no dose-effect relation observed. The concentration might be too low to observe this effect. The MAC-MKN is set on 0.48 μg/L, because other organisms are harmed by metazachlor at an exceedance of this concentration (Vonk, 2013). M. spicatum is not the most sensitive organism to metazachlor. The EC50 of Myriophyllum heterophyllum to metazachlor is 80 μg/L, which is not achieved (Vonk, 2013). The EC50 of M. spicatum is assumed to be approximately the same as the EC50 of M. heterophyllum. It is possible that the effect of metazachlor is visible when M. spicatum is exposed to a higher concentration, especially since the matched pore water concentration turns out to be lower than expected with the accessory sediment concentrations.

The plants from conditions norm, 5x norm, 20x norm were grown in the same aquaria. 200x norm was grown in 2 other aquaria, of which of those two aquaria had the same volume as one aquarium of one of the other conditions (see Appendix f, photos a). However, the 2 aquaria of 200x norm, is more squared while the aquaria of the other conditions are more rectangular. So if the metazachlor treatment was not harmful in any of those condition, the fact that the plants had more space in 200x norm, could have played a part in the fact that the plants of 200x norm showed more growth than plants of the other conditions. This could explain why the plant of 200x achieved a better result, against the expectations.

The reference that was used for the spiked sediment is the same reference that was used for the bioassay. The reference test is not run at the same time and grown in cores instead of aquaria. Besides,

(22)

the reference contained one plants per cores, while the spiked sediment test contained at least 2 plant per aquaria. Because the surface of the cores where the reference replicates are grown was smaller, the height of the sediment cores was taller. To create a more reliable reference, the test with the reference should have run at the same time, in similar aquaria. Therefore, the reference is less reliable than desired.

One of the purposes of the bioassay was to asses to ecological quality of the agricultural location. It was expected that the sediment of the agricultural site would result in more negative effects on M. spicatum than the agricultural water. Results of the bioassays support this hypothesis, which indicates that the quality of the sediment is poor, and the quality of the water is good. Performing toxicity tests of other organisms with the water and/or sediment of the agricultural site might be helpful to assess the ecological quality even more. The agricultural sediment affected the root/shoot ratio of M. spicatum. However, it is not clear which compounds are responsible for this effect. Further research could include chemical analysis, which give more information about the chemical composition of the water and sediment of the agricultural site. Metazachlor was an example of one compound that could potentially be responsible for this. Test results failed to prove this. To be able to better investigate the impact of metazachlor on M. spicatum with spiked tests, the Koc and Kow have to be determined experimentally. More research on how metazachlor acts in sediment needs to be done, since the chemical properties influence the uptake, translocation, accumulation and elimination of metazachlor (Diepens et al., 2014). Besides, due to the small sample size of each concentration of the spiked sediment test, of which the standard deviations are quite large, a confirmative study with more replicate is needed before application into ecotoxicological guidelines.

Overall, the results of this research show that M. spicatum is affected by the sediment quality. Since macrophytes such as M. spicatum are of great importance for freshwater ecosystems, the impact of poor sediment quality deteriorates the condition of the ecosystem (Knauer et al., 2006; Nuttens et al., 2016). To maintain good water quality, sediments need to be of good condition (Brinke, 2015). Because sediment quality is of high concern, they need to be assessed when it comes to ecological risk

assessment.

6. Conclusion

The aim of this research was to investigate whether agricultural sediment and water hampers the development of the rooting aquatic macrophyte M. spicatum. In phase one there is investigated whether contaminated water and sediment from the agricultural site hampers the development of M. spicatum, and does the contamination in the water or sediment affect the macrophyte more? In the second phase, there is investigated to what extent the herbicide metazachlor in water and in sediment, detrimentally affect Myriophyllum spicatum? M. spicatum was affected by the sediment of the agricultural site. In conclusion, contaminations in the sediment of the agricultural site in Almere Buiten hampers the development of M. spicatum. However, the water of this location is not limiting the development of this plant. The quality of the water is controlled; however, sediment is not measured whatsoever. This experiment shows that the impact of the sediment is bigger than the impact of the water, so to maintain a healthy ecosystem, a shift needs to be made towards measuring and maintaining non-toxic sediment. Therefore, sediments need to be assessed to determine the quality of freshwater ecosystems. Results were inconclusive, whether metazachlor is more harmful to M. spicatum when it is found in the sediment or in the water. Besides, it did not become clear to what extent metazachlor hampers the development of M. spicatum. Characteristics of metazachlor need to be investigated more to create a

(23)

more reliable toxicity test for both sediment and water before the toxicity of metazachlor in sediment and water in relationship to M. spicatum can be determined.

7. References

Adams, W. J., Kimerle, R. A., & Barnett, J. W. (1992). Sediment quality and aquatic life assessment. Environmental Science and Technology, 26(10), 1864–1875.

Ahrens, L., Daneshvar, A., Lau, A. E., & Kreuger, J. (2015). Characterization of five passive sampling devices for monitoring of pesticides in water. Journal of Chromatography A, 1405, 1-11.

Atlas Bestrijdingsmiddelen in Oppervlaktewater (2019). Derived on 14-06-2019 from http://www.bestrijdingsmiddelenatlas.nl/

Australian Pesticides and Veterinary Medicines Authority (2016). Public release summaries: the

evaluation of the new active metazachlor in the product Butisan Herbicide. Derived on 12-06-2019 from https://apvma.gov.au/sites/default/files/publication/26306-prs-butisan-herbicide.pdf

Babut, M. P., Ahlf, W., Batley, G. E., Camusso, M., de Deckere, E., & den Besten, P. J. (2005). International overview of sediment quality guidelines and their uses. In Use of sediment quality guidelines and related tools for the assessment of contaminated sediments/Wenning, Richard J.[edit.] (pp. 345-381).

Borja, A., Valencia, V., Franco, J., Muxika, I., Bald, J., Belzunce, M. J., & Solaun, O. (2004). The water framework directive: water alone, or in association with sediment and biota, in determining quality standards?. Marine Pollution Bulletin, 49(1-2), 8-11.

Brinke, A., Buchinger, S., Reifferscheid, G., Klein, R., & Feiler, U. (2015). Development of a sediment-contact test with rice for the assessment of sediment-bound pollutants. Environmental Science and Pollution Research, 22(16), 12664-12675.

De Baat, M. L., Bas, D. A., van Beusekom, S. A. M., Droge, S. T. J., van der Meer, F., de Vries, M., ... & Kraak, M. H. S. (2018). Nationwide screening of surface water toxicity to algae. Science of the Total Environment, 645, 780-787.

Diepens, N. J., Arts, G. H., Focks, A., & Koelmans, A. A. (2014). Uptake, translocation, and elimination in sediment-rooted macrophytes: a model-supported analysis of whole sediment test data. Environmental science & technology, 48(20), 12344-12353.

Google Maps. Derived on 14-06-2019 from

https://www.google.com/maps/place/Kottertocht/@52.419053,5.2682184,13z/data=!3m1!4b1!4m5! 3m4!1s0x47c618075da86015:0x57bc3ca240fbe332!8m2!3d52.4190058!4d5.3032378

Knauer, K., Mohr, S., & Feiler, U. (2008). Comparing growth development of Myriophyllum spp. in laboratory and field experiments for ecotoxicological testing. Environmental Science and Pollution Research-International, 15(4), 322.

McCann, J. H., Greenberg, B. M., & Solomon, K. R. (2000). The effect of creosote on the growth of an axenic culture of Myriophyllum spicatum L. Aquatic Toxicology, 50(3), 265-274.

(24)

Nuttens, A., Chatellier, S., Devin, S., Guignard, C., Lenouvel, A., & Gross, E. M. (2016). Does nitrate co-pollution affect biological responses of an aquatic plant to two common herbicides?. Aquatic Toxicology, 177, 355-364.

OECD (2014a). Test No. 239: Water-sediment Myriophyllum spicatum toxicity test. Derived on 10-06-2019 from: https://www.oecd-ilibrary.org/docserver/9789264224155-en.pdf?

expires=1555003592&id=id&accname=guest&checksum=D9BAAAB859122B6CB848D5813C544AF5 OECD (2014b). Test No. 238: Sediment-Free Myriophyllum spicatum toxicity test. Derived on 10-06-2019 from: https://www.oecd-ilibrary.org/docserver/9789264224155-en.pdf?

expires=1555003592&id=id&accname=guest&checksum=D9BAAAB859122B6CB848D5813C544AF5 PPDB: Pesticide Properties DataBase (2018). Metazachlor (Ref: BAS 47900H). Derived on 12-06-2019 from https://sitem.herts.ac.uk/aeru/ppdb/en/Reports/450.htm

R Core Team (2018). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria.

Roex, E., Klein, J., Weert, J. D., Zelfde, M., & Tamis, W. L. M. (2016). Rapportage en evaluatie Landelijk Meetnet Gewasbeschermingsmiddelen 2014.

Rueden, C. T.; Schindelin, J. & Hiner, M. C. et al. (2017), "ImageJ2: ImageJ for the next generation of scientific image data", BMC Bioinformatics 529,

Salomons, W., De Rooij, N. M., Kerdijk, H., & Bril, J. (1987). Sediments as a source for contaminants?. Hydrobiologia, 149(1), 13-30.

Schonevijver.nl (2019). Derived on 27-06-2019 from https://www.schonevijver.nl/index.html Tunić, T., Knežević, V., Kerkez, Đ., Tubić, A., Šunjka, D., Lazić, S., ... & Teodorović, I. (2015). Some

arguments in favor of a Myriophyllum aquaticum growth inhibition test in a water–sediment system as an additional test in risk assessment of herbicides. Environmental Toxicology and Chemistry, 34(9), 2104-2115.

Vethaak, A. D., Hamers, T., Martínez-Gómez, C., Kamstra, J. H., de Weert, J., Leonards, P. E., & Smedes, F. (2017). Toxicity profiling of marine surface sediments: a case study using rapid screening bioassays of exhaustive total extracts, elutriates and passive sampler extracts. Marine Environmental Research, 124, 81-91.

Vonk, J. W., Smit, C. E., & de Jong, F. M. W. (2013). Environmental risk limits of metazachlor in water: A proposal for water quality standards in accordance with the Water Framework Directive. Derived on 12-06-2019 from https://rivm.openrepository.com/handle/10029/311506

Wester, P. W., & Vos, J. H. (2003). Identification of endocrine disruptive effects in the aquatic environment-a partial life cycle study in zebrafish. Derived on 12-06-2019 from

(25)

8. Appendix

8.1 Appendix a: Maps and photos of sampling locations

The maps of the agricultural location and the reference location are displayed (Figure 7; Figure 8). In the map of the agricultural site the greenhouses are indicated.

Figure 7. The map of the location Kottertocht (Google Maps, 2019). The pin is set on the point where the samples are taken from. The tractors indicate the multiple greenhouses at both site of the Kottertocht.

(26)

Figure 8. Location Science Park (Google Maps, 2019). The pin is set on the point where the samples are taken from.

8.2 Appendix b: The composition of Dutch Standard Water

The Dutch Standard Water contains 2 mL per liter DSW of following suspension:

- 100 g/L CaCl2 · 2H2O - 90 g/L MgSO4 · 7H2O - 50 g/L Na HCO3 - 10 g/L K2CO3

(27)

8.3 Appendix c: Use of ImageJ

Total shoot length and total root length are measured using ImageJ (Figure 9). Photos used for these measurements were scaled. The tool ROI Manager is used to draw multiple lines at the same time, of which the length of these lines could be summed, to obtain the total length.

Figure 9. Screenshot of measurement of total shoot length and total root length. a) Total shoot length is measured using the tool ROI Manager. b) Total root length is measured. The roots are detachted from the plant in orde to be able to make a picture where all the roots are visible. Lines are drawn from the beginning to the end of the root.

a)

(28)

Appendix d: Abiotic measurements

Measurements of pH, voltage, dissolved oxygen concentration, oxygen saturation, conductivity and mean temperature are shown (Table 3). For the bioassay sediment and for spiked sediment,

measurements took place weekly, where for the bioassay water measurements took place twice a week. Some values are not reported properly and got lost, that is why some values are missing. No extreme deviations are shown; therefore, the abiotic conditions are considered as stable.

Table 3. Abiotic measurements of the bioassays and the spiked sediment test.

Bioassay sediment, agriculture site

Dag 0 Dag 7 Dag 14 Dag 21 Dag 28

pH 8.1 7.9 8.0 7.9 8.4

Voltage (mV) -74.2 -63.5 -71.6 -63.3 -95.1

Dissolved oxygen concentration

(mg/L) 9.1 7.3 7.1 6.6 8.5

Oxygen saturation (%) 100.4 80.0 70.1 90.0

Conductivity (µS/cm) 668.0 955.0 1041.0 715.0 1077.0

Mean temperature* (degree

Celsius) 19.8 19.0 18.6 18.5 18.9

Bioassay sediment, reference

pH 7.8 8.0 8.1 8.5 8.3

Voltage (mV) -57.2 -69.1 -77.5 -96.4 -87.7

(mg/L) 8.8 8.1 8.1 8.4 8.4

Oxygen saturation (%) 96.0 88.0 88.5 92.2 90.0

Celsius) 20.0 19.1 19.0 19.1 19.2

Bioassay water, agricultural site**

Dag 18 Dag 21 Dag 25 Dag 28 pH 7.9 8.5 8.5 8.4 8.5 8.4 8.4 8.4 8.4 Voltage (mV) -64.8 -100.9 -98.2 -92.0 -96.5 -92.0 -91.4 -94.5 -92.6 Dissolved oxygen concentration (mg/L) 10.7 9.4 8.8 7.1 8.5 7.9 9.2 9.9 8.8 Oxygen saturation (%) 100.3 100.8 95.4 77.0 94.4 82.9 99.7 107.6 95.3 Conductivity (µS/cm) 3320.0 3400.0 2287.8 3133.3 3430.0 3430. 0 3410. 0 3373. 3 3380.0 Mean temperature* (degree Celsius) 11.5 19.0 18.9 19.2 19.3 19.0 19.4 19.1 19.4

(29)

Bioassay water, reference**

Dag 18 Dag 21 Dag 25 Dag 28 pH 7.6 8.5 8.5 8.4 8.4 8.4 8.5 8.2 8.5 Voltage (mV) -49.3 -96.0 -97.4 -83.8 -91.1 -95.1 -98.3 -83.6 -99.3 Dissolved oxygen concentration (mg/L) 9.7 8.1 7.0 6.6 6.8 7.3 6.9 7.9 7.7 Oxygen saturation (%) 87.2 85.7 76.0 70.0 74.7 76.5 73.9 86.3 83.3 Conductivity (µS/cm) 1780.0 1764.7 1794.0 1813.7 1790.3 1701. 0 1807. 7 1544. 0 1770.0 Mean temperature* (degree Celsius) 10.3 19.3 19.1 19.1 19.2 19.2 19.0 19.5 19.4

Spiked sediment test**

pH 7.5 7.7 8.0 8.1 8.2

Voltage (mV) -44.9 -54.8 -70.8 -73.7 -81.1

(mg/L) 5.7 5.3 7.8 8.6 8.1

Oxygen saturation (%) 59.5 55.6 81.5 91.2 87.5

Celsius) 19.2 18.1 18.3 18.4 18.3

* The mean temperature is based on the temperature measurements of the oxygen meter, conductivity meter and pH meter

** For both water tests: All the measurement are the mean of the measurement of three replicates For the spiked sediment test: All the measurement are the mean of the all the aquaria with different concentration (one aquarium with norm, one aquarium with 5x norm, one aquarium with 20x norm, two aquaria with 200x norm

(30)

Appendix e: Photos of visual findings

Multiple organisms are found during the test (Figure 10). Not every organism is identified. Besides, during the experiment, algae growth was discovered on multiple places

Figure 10. Photos of visual findings. a) Prior to experiment, this organism was found in the large plastic core, containing the agricultural sediment core of sample 8. This is presumably a Asellus aquaticus, multiple animals of this species were

found in the agricultural cores. b) Multiple organisms are present. Worms are crawling out of the sediment. On the plastic cores are organisms present that are presumably Asellus aquaticus. Picture made on 25-04-2019, on day 9 of the sediment bioassay. c) Algae growth on plug of the water bioassay. Picture made on 06-05-2019, on day 11. d) Presumably scum of algae on the side of the plastic cores of the water bioassay. Picture made on 23-05-2019, on day 28. Cores number 13 and 14 had the most severe scum.

c)

b)

a)

(31)

On top of the water in the aquarium of the bioassay, agriculture, a film of presumably algae had arised (Figure 11). This layer was spotted first on day 13 of the experiment (29-04-2019). This layer could have lowered the light intensity that reached the plants.

Figure 11. Green/brownish layer on the water of the agricultural sediment bioassay. Picture made on day 18 of the experiment (04-05-2019).

(32)

Appendix f: Spiked water test

The spiked water test was failed. This is mostly due to the use of flowering plant and contamination of the samples. As a result of the contamination, water became cloudy, a lot of insoluble particles were found in the water, plants were entangled and died (Figure 12).

Figure 12. Contamination of the spiked water test. a) Rows left to right: Norm, 5x norm, 20x norm and 200x norm. The water of each jar is cloudy. Picture made on day 16 of the experiment (23-05-2019) b) Spiked water test, sample 20x norm 1. Picture made on day 16 of the experiment. c) Due to what is presumably algae growth, the plant is entangled and almost not visible anymore. Sample 5x norm 4. d) Plant that is considered dead; 200x norm 5. Contamination of the plant is visible.

Assessing contaminated agricultural sediment and water quality with the macrophyte Myriophyllum spicatum