- Good afternoon. It is my pleasure to welcome you to our online version of Victoria's third Environmental Science Series event of 2021. Today's talk will be focusing on the effects of pharmaceuticals in aquatic environments, bridging the gap between lab and field. My name is Professor Mark Taylor. I'm Victoria's Chief Environmental Scientist at the Environment Protection Authority in Victoria. I would like to begin by acknowledging the Aboriginal people as the First Peoples and Traditional Custodians of the land and water on which we live, work and depend. I am speaking to you today from the Dharug country in Sydney, Australia. We pay respect to Aboriginal Elders, past and present, as Victoria's environmental regulator, we pay respect to how Country has been protected and cared for by Aboriginal people over many tens of thousands of years. We recognise the unique cultural and spiritual significance of land, water and all that is in the environment and the continuing connection and aspirations for Country of Aboriginal people and traditional custodians.

So just a few notes with respect to this livestream event. Please be patient with respect to any technical difficulties. The structure of today's event will feature on an introduction by myself followed by, most importantly, presentations by guest speakers and there will be some time for questions and answers and a brief conclusion. The event is recorded so if you need to leave early or if you miss any parts of this session, you can watch it later via EPA's website. The session has live closed captions to make it accessible to all. We will open the questions and answer section now, but respond after the presentations. Please provide your questions at the icon with the question mark on your screen. We will try to get as many questions in as possible and try to get them answered today, and if not, they will be responded to later via email.

Please provide your contact details so we can respond directly to you. To today's conference, we're here to talk about pharmaceuticals and their effects on aquatic environments and the work underway to bridge knowledge gaps between field and laboratory studies. For the purpose of our discussion, pharmaceuticals can be defined as any product used for personal health, cosmetic reasons or used by agribusiness to enhance growth and health.

Without doubt, pharmaceuticals have undeniable benefits. They have enhanced food production, but there are difficulties in addressing their potential impacts to our environment and human health. Pharmaceuticals are considered to be an emerging contaminant of concern. This means there are a variety of synthetic pharmaceuticals in the environment, there is limited or absent information or full information about their concentrations in the environment, sources, spatial distribution and fate. The risk they pose to our health and the environment is not fully understood, therefore information about the toxicity and risks associated with pharmaceuticals and the environment and our consumption of them inadvertently is important. With the growing and ageing human population, pharmaceutical use has increased globally. This means we have a greater number of pharmaceuticals passing through our bodies, wastewater systems and into the environment. Global research has shown that pharmaceuticals are found in surface and groundwaters that are used in irrigation as well as drinking water.

Previous studies to date have shown that pharmaceuticals can disrupt sensory capabilities of animals and their ability to locate food and mates. In addition, exposure to pharmaceuticals can alter the ability of animals to capture pray or escape predators. One of our guest speakers today will be elaborating further on these effects. Pharmaceutical challenges traditional water quality management because it requires new technologies in wastewater treatment and different behaviours in industry, agriculture and health sectors along with society in order to address the problem. Further work is needed to better understand background concentrations of pharmaceuticals in the environment and the risk that they pose to human health and the environment. The role of the EPA is to protect the environment and human health from risks of harm associated with pollution and waste. The EPA has started examining baseline data for a number of emerging contaminants in Victoria as we will hear today. Our work is supporting this particular area of key research involving the following items: what are the effects on our environment and on humans; what risks do they pose and how can we minimise those risks? EPA continues to work with other regulators, research partners and industry both in Australia and internationally to build knowledge and understanding of pharmaceuticals, so today in discussing this important topic and the effects on the environment and public health, it helps us to build our knowledge and understanding of the problem.

On that note, it is with great pleasure that I introduce our two guest speakers for today, Dr Minna Saaristo and Professor Tomas Brodin. Dr Minna Saaristo is a senior applied scientist and a leader of the Emerging Contaminants Program at EPA. Minna has led projects assessing background concentrations of emerging contaminants in Victoria, exploring effects of wastewater treatment plant effluent on wildlife and unravelling the presence of emerging contaminants in recycled water. Minna's expertise is in behavioural ecotoxicology and she has over 16 years of international and national experience on assessing the impact of chemical contaminants on wildlife.

Also, I am absolutely delighted to have our international guest speaker, Professor Tomas Brodin from Sweden who works at the University of Agricultural Sciences to talk to us about his research in this space. During the last decade, Tomas has focused on studying the ecological effects of pharmaceuticals in aquatic systems. He has a background in evolutionary and behavioural ecology and is now using this knowledge to increase the ecological relevance of chemical risk assessment in general, but also specifically related to pharmaceuticals. His research bridges the gap between lab and the real world by combining his laboratory experiments where a mechanistic understanding with respect to large scale field studies that help answer the big question what happens in the lake and stream when pharmaceuticals are released into those environments.

So, today, I look forward to listening to you both and thank you for both of you for joining us today and preparing your talks which I know how much that takes and how much effort. You'll have gone over your talks and slides multiple times. I will now hand over to Minna who will provide an overview of the EPA's emerging contaminants program. Thank you very much, Minna.

- Thank you, Mark. Hello, everyone. It's wonderful to see so many of you out there today. I am very excited to be here today and talk about a topic that I'm very passionate about. Some of you in the audience might know that I have researched pharmaceuticals for I would say the 16 years that I've been in this space, and today I'll be sharing the work that I've done here at the EPA Victoria.

Before I move into pharmaceuticals I take a few minutes to talk about the Emerging Contaminants Program that we have at EPA Victoria which some of you might not be aware of. The program aims to monitor and assess the quality of water, land and air in a context of emerging contaminants and aims to identify and assess environmental and human health risks associated with these contaminants, but most importantly, it's there to guide duty holders to reduce risk and prevent harm to human health and the environment and also to increase EPA’s understanding of emerging contaminants so we can develop better environmental standards and guidance for management.

We're here to talk about pharmaceuticals and I'm sure we all agree it is a growing global problem. Let's put some numbers on the problem. The global pharmaceutical market has been estimated to exceed US 1.5 trillion dollars by 2023 which is actually an annual growth rate of 6 per cent over the next five years. Bringing the numbers back to Australia, in 2017/18 there were more than 300 million prescriptions dispensed in Australia alone. This is an increase of 1.5 per cent. If you look at the drugs listed on the left, those are the 10 most prescribed drugs in Australia and one of them here in the middle is Cefalexin with is an antibiotic and I'll be talking about it more later today.

It's important to be aware of where actually are these pharmaceuticals coming from. It's easy that we think about the human pharmaceuticals, but they also come from veterinary practices, such as agriculture, companion animals and livestock and all the human pharmaceuticals resources and veterinary pharmaceuticals, all these together end up in freshwater environments all the way to the groundwater and even in terrestrial ecosystems and therefore it's not surprising that more than 4,000 active pharmaceutical ingredients have been detected in the environment.

What makes this pharmaceutical pollution especially an emerging concern is that these chemicals have been designed to be really highly active and really interact with receptors in humans and animals. Because we're using them all the time, they’re actually continuously released into the environment which means the non‑target organisms are exposed to these for a very long period of time. One element that is very specific for pharmaceuticals is that they actually have something called non‑monotonic dose response. The response may be greater at the lower dose than the higher dose. This means low concentrations really matter.

Let me walk you through work that I was lucky to lead here at EPA Victoria. We did the sampling last year in March 2020. The aim of the study was to pretty much start with the basics. Let's establish background concentrations of pharmaceuticals and personal care products, PPCP, and endocrine disrupting chemicals, EDCs, in fresh water and in biota across different sites in Victoria. We wanted to investigate, is there a difference between the upstream and downstream sites and is there also association to the discharge points?

Let's look at those sites a bit more closely. Here on the right, you see a map of the sampling sites. Overall, we covered 18 different sites. There is ‑ the blue reference site, we have three reference sites, two along Lerderderg River, one along the Watts River, and then we sampled Burrumbeet Creek, Campaspe River, Jacksons Creek, Merri Creek and Mullum Mullum Creek. We had three sites per location, the yellow here is upstream, then there's a hotspot and downstream. The upstream site was approximately 2 kilometres from the hotspot and the downstream site was another two kilometres from the discharge point. The discharge point is not effluent ‑ we didn't collect the effluent, we collected what's in the waterway, so in a creek or river. It's an environmental sample. From each site, we collected both fish and water, so approximately aim to have 18 fish per site, but the guys actually were very successful with catching fish so I would say we had more than 20 individuals from each site. Overall, we're talking about sample size of more than 300 fish that we were able to analyse through this sampling campaign.

For water analysis, we used two methods. We used spot sampling and then passive sampler. Here on the bottom left, you see this cannister that has inside filters and those filters were deployed for 28 days. You will see in my talk today I talk about day zero and day 28. So the day zero is when the passive samples were deployed and day 28 is when they were retrieved. We have three data points from each site.

Both fish and waterway analysed for pharmaceuticals, personal care products, EDCs and also for PFAs, but today we'll be focusing on the pharmaceuticals and EDC results.

What did we find? Here is pharmaceuticals in water data based on collection from the day 28, and on the Y axis there, it's total pharmaceuticals from zero to 10 micrograms per litre, and on the X axis we have different sampling sites. Left to right, first are the reference sites, then we have Burrumbeet Creek, Campaspe, Jacksons, Mullum Mullum and Merri, and we have these three sites for location to the side here. There's upstream, discharge and downstream. I've listed here on the right the top 10 pharmaceuticals that were detected most frequently. Starting from ‑ I doubt that not everyone in the audience know what these are so I'm going to quickly go through these. Venlafaxine is an antidepressant, Paracetamol is a painkiller, Oxazepam is a medication used to treat anxiety disorders, Methamphetamine is a recreational drug, Gabapentin is medication used to treat seizures, Furosemide is medication used to treat high blood pressure, Diclofenac is an anti‑inflammatory drug, Cefalexin is an antibiotic, Carbamazepine is medication used to treat epilepsy and pain and Acesulfame K is an artificial sweetener. So a selection of painkillers, anti‑depressants and high blood pressure medication. These 10 are quite standard, what have been detected elsewhere in the world as well.

Let's look at more closely the differences because that was one of the keys that we wanted to see differences between locations as well. It's not surprising that the discharge point has the highest concentrations and if you look at the Burrumbeet Creek, the discharge point has up to 10 micrograms per litre compared to Jacksons Creek which has up to 2.83 micrograms per litre, but what is interesting is people often think when the pharmaceuticals end up in the environment, they get diluted and you hardly see anything in the downstream. If you look at the Burrumbeet Creek site, downstream was 2.4 km for the discharge, we still get three micrograms per litre concentrations.

What other differences between these locations are things like they have different treatment levels, so at the Burrumbeet Creek, they use secondary treatment, while in the Merri Creek, they use tertiary treatment, so that's the obvious difference ‑ makes a difference for pharmaceuticals. The other thing is what kind of waste they are receiving. Burrumbeet Creek site is receiving both domestic and industrial waste, but also at the Burrumbeet Creek site, the mean flow per day is more than 8,000 kilolitres per day while in Merri Creek, it's around 2,600. So then the flow, waste and treatment all makes a difference when we look at what are the pharmaceuticals coming out the other end.

Let's have a look at what did we find in terms of different collection methods. I said we used two different methods for water. We used spot sampling and passive samples and here on the Y axis we have mean concentration of pharmaceuticals in nanograms per litre from zero to 1,200 and on the X axis we have the different pharmaceuticals. We analysed 72 different pharmaceuticals and personal care products. What we found was that out of the 72 with a passive sampler, we were able to detect 59, while with the spot sampling method we collected 39. Again it seems pretty obvious that when you have a filter that filters the water for all the time, you would get more, but it's good to show evidence that yes that was the case in this study. But also what is important to note that these two different methods actually then show slightly different results if we look at then what were the top three chemicals that were coming out as being the top chemicals of concern.

So with the passive sampler, the Venlafaxine was the one detected at highest, then came Carbamazepine, Tramadol and Oxazapam. With the spot water samples, it was actually Cefalexin that was the highest, and then came Furosemide, Hydrochlorothiazide which is a high blood pressure medication and then Venlafaxine. So Venlafaxine came out in both of them, but then the Cefalexin antibiotic was clearly detected more commonly in the spot samples. Again, the take‑home message is it's important to use multiple different methods when you're monitoring for pharmaceuticals in water.

Moving on to what did we find in terms of pharmaceuticals in biota. As I mentioned, we did have a sample size of more than 300 fish, but because the summary, due to time constraints, I'm only going to show data for fish we were able to measure the edible portion. These fish are all big enough that for the analysis, you need a certain volume. If the fish is small, you have to analyse the whole fish. This is a summary of data for black fish, carp, eel, redfin, and then there’s the crustacean, yabbies, as well that we got from the reference site. We analysed 22 different pharmaceuticals, so unfortunately not as many as we were able for water because the method is still in development for pharmaceuticals, but out of the 22, only three were above the detection limit. That's always a bit disappointing when you really want to see what is out there because it doesn't ‑ even if we only detected three, it doesn't mean that the rest of them are not there. So these were the ones that ‑ the concentrations were high enough that we were able to see. What was interesting, Venlafaxine was coming out as the highest concentrations out of all these.

Let's look at the Venlafaxine, which was anti‑depressant, a little more closely. Here I've summarised this is the data based on what was the Venlafaxine concentration in fish muscle tissue across all sites. The first panel was reference sites then we have upstream discharge and downstream sites and on the Y-axis we have Venlafaxine ... from zero to 150. On the X axis, we have different species from black fish to yabby and the same as I showed in the previous slide.

What was clear was the Venlafaxine was ‑ the highest concentration was at the discharge point and the concentrations were from limit of reporting which was 5 micrograms per kilogram up to 150. It was only detected in redfin perch. When we look at especially the 150 microgram per kilogram, the result was a bit of a surprise and we went back to the lab and checked if that was a real result, and yes it is. But when we then start looking at the data, it was an individual that was on one of the biggest ones that we caught, animals, a fish that weighed 448 grams and it was 310 centimetres. So it was a decent size redfin. When we look at the waterway where we caught that from, actually from the downstream site, the fish were ‑ the redfins that we caught were showing Venlafaxine as well and we caught ‑ so at the discharge point in that waterway, we caught three redfins, two had detectable levels of Venlafaxine, we caught four from downstream and all of them had more than 25 micrograms per kilogram in the edible portion of their tissues and upstream none of them - we caught 7, and none of them had. Clearly in this waterway, fish are exposed to Venlafaxine.

Just a reminder what were the water levels because always we need to ‑ traditionally, monitoring has been done based on water sampling. So if we look at the waterway where the fish had such high concentrations, you would be looking at these yellow boxes. Therefore, the discharge point would have - mean concentration was .5 micrograms per litre which is high, but you might actually show any red flags at all, but looking at water levels .5 micrograms per litre, fish have up to 150 micrograms per kilogram, definitely take‑home message is it's important to analyse both water and biota to have a better understanding of the risks of these pharmaceuticals.

Then to quickly go through, we did the traditional ‑ we look at bioaccumulation factors based on the levels detected in the water and in tissues and yes, there's evidence for bioaccumulation for Venlafaxine and Carbamazepine. Then my colleagues at the environmental public health unit did margin of exposure calculations and based on their calculations, there seems to be a relatively low public health risk for consumption of Venlafaxine and Carbamazepine. Despite that, I think it’s important we are aware of the growing body of research that shows that exposure to Venlafaxines have adverse effects on non‑target organisms and there's only four papers that I'm showing here, but there's tens and tens that show that exposure to Venlafaxine changes the brain serotonin levels in fish, high bioaccumulation, reduces white-bream larvae, survival of those larvae reduces embryo production and most importantly, it is changing the behaviour of these organisms as well. Yes, some of those concentrations are relatively high like 100 micrograms per litre, then there is the Galus et al. study that exposed zebrafish for 5 micrograms per litre showing reduced embryo production. Definitely, the levels we are already measuring the environment - there is evidence of bioadverse effects.

This was something that I discovered when I started to look into Venlafaxine, that actually ‑ there is evidence that the current treatments are not really treating Venlafaxine. There's really low removal efficiency and there are studies by colleagues from CSR in Adelaide that actually show that effluent concentrations were up to seven times higher than the influent concentrations when they were measuring Venlafaxine. It looks like there's a possibility that the transformation product is converting back to the parent compounds during treatment.

All things considered, I would say there's more work to be done and looking at future directions for EPA, definitely ideally we need a more strategic, but a longer term monitoring plan that we can go back to the same sites and look at ‑ go back to the same sites, but expand the network and look at what else is out there. There's a lot of things that we didn’t measure, we didn’t look at levels in plants, in birds, in invertebrates, and we didn't look at sediment as well. So there's endless knowledge gaps still there for us to conquer. To answer all those research questions and all the knowledge gaps, the key is then to collaborate and create partnerships to really get a better handle on what are the risks of these pharmaceuticals. What we can do, we need to be more active in developing guidelines and the environmental reference standards for pharmaceuticals and what we also can do, what we're doing today, we can start communicating and educating decision-makers of these potential risks of pharmaceuticals.

As a nation, what we should be moving towards doing is promoting more whole lifecycle approach to pharmaceuticals, what European Union is already doing, looking at actually designing better products and designing more ‑ promoting greener manufacturing and really questioning do we really need to use all those pharmaceuticals and are we using it for the right reasons? Then when we do use them, let's make sure that we collect them properly, we dispose of them properly and then when they end up in the wastewater treatment plant, then they are treated to level that causes minimal harm to the environment. To be able to do all that, we do need to develop better chemical strategies and action plans to really tackle pharmaceutical pollution. That was the end of my talk. Thank you. Now I hand over to my dear colleague, Tomas. Over to you.

- Thank you very much for inviting me to share with you some of the work I've been doing for the last 15 years in the realm of pharmaceuticals in the environment and the ecological effects thereof. It's a pleasure to be here. It's a bit early in the morning here in Sweden, but the early bird catches the worm, they say, and maybe I will today. I will give you a short introduction of pharmaceuticals in the environment, and ‑ let's see if the takeover of the ‑ seems to not work at the moment. Maybe now. Yep. There we go.

Perfect. Not perfect, but good enough.

The title of my talk is the Effects of pharmaceuticals in the environment - bridging the gap between lab and field. When I started working on ecological effects of pharmaceuticals in the environment about 15 years ago, we knew virtually nothing about what potential effects were going on in the wild. We didn't know what pharmaceuticals were out there, even. Some studies had been done, but we were pretty much fumbling in the dark in the beginning, and especially as an ecologist, trying to figure out what drugs, for example, to look for and to analyse to expect effects of is really hard. It was a stroke of luck that I held a talk at a university in Sweden and one of my colleagues heard about my studies of animal behaviour and animal personality as I was working with at that time and when he was driving home from the university that night, he heard a chemist speaking on the radio, Swedish science radio telling everyone that he's found 26 different pharmaceuticals in rainbow trout exposed to wastewater. This colleague of mine, he sent me a text, this guy found pharmaceuticals in fish and you were talking about pharmaceuticals and how they might affect fish behaviour, because basically my background is in evolutionary and behavioural ecology, and so I set up a meeting with this chemist that I never heard of, never seen, and it turned out he was sitting three metres above me in the same house, which just shows how airtight the walls between chemistry and ecology can be, even within the university, but we set up a meeting in a room with no windows, it was really cold, the three of us, the guy who also was driving the car and listening to the radio, he was an environmental scientist, and also a good web specialist. We sat down, the four of us, and tried to design the first experiment. I'll come back to this first experiment a little bit later in my talk.

I was going to give an introduction to why pharmaceuticals are so special and important, but Minna beautifully presented that to you, but I'll briefly run through it anyway. During my 15 years of studying pharmaceuticals, it's really been a huge increase in the awareness that pharmaceuticals might actually be bad for us and that they are exposing wildlife globally. When I started, the idea was that pharmaceuticals are good. They are designed to do good for us, why should they harm the environment? I mean that's a rationale behind that statement and that belief, but as we now know, there are pharmaceuticals that endanger wildlife and ecosystem functioning and even human health, and these are being excreted and exposing humans and animals all over the world. As Mark said, we go through our slides many times, but as you can see down on this slide, it still says Spain and Sweden there. Because most countries have little to no regulation when it comes to pharmaceuticals, they are not handled in a chemical act in Sweden because they are of utmost importance to human health so they've been excluded. So there are basically no rules or regulations of how much pharmaceuticals you can pollute, and I will come back to why that was a good thing for us at one point later.

Okay. So pharmaceuticals are increasing. We heard about that in Mark's introduction. I'm going to show you in relation to other increasing things in the world, how much pharmaceuticals actually are increasing. This is atmospheric carbon dioxide increase. We know about the increase ‑ a lot about it in the radio and TV and media, and it has tremendous effects globally of course. Human population is also increasing and this is between 1955 and 2015, and human increasing population is also affecting the world tremendously. But if we look at the chemical industry output which is here, see there's a huge proportional change since ‑ especially since 1996/7, but all the way back to 1960s increasing a lot, the chemical industry output. Then if we look at pharmaceuticals, it's an enormous increase from 1995 to today, or to 2015, and from 2015 until today, we have another 24 per cent increase in pharmaceutical consumption. There is a huge burst of pharmaceutical consumption and of course, these pharmaceuticals end up in the environment, or at least many of them do and potentially affects wildlife.

So needless to say, we are living in a medicated world. The pharmaceutical pollution is a global issue and we use a lot, 4.5 trillion doses approximately per year. There are five to 6,000 increasing daily almost products on the market, and over 600 pharmaceuticals have been detected in the environment today. It's actually over 700 now. And it's been reported across 71 countries. This has actually increased a lot as well since Alistair Boxall's recent study he submitted, just about published - it’s coming out within weeks, where he looked at water from over 100 countries and there are a lot of new world records when it comes to concentrations of pharmaceuticals in the environment in that paper. Scary, but not unexpected.

So when we think about how pharmaceuticals enter the environment, this is often the way, through wastewater. We have a bunch of different types of pharmaceuticals in these waste waters. These are a few of the groups that we have either been looking at the ecological effects of or bioconcentration of in aquatic wildlife. At the moment, we're working mainly with anti‑anxiety drugs and fiddling a little bit with antihistamines and antibiotics.

Why are pharmaceuticals different from other chemicals? Well, one important reason is that they are designed to have biological effect at low doses. So when you do ecotoxicological tests on pharmaceuticals, you're looking for toxic effects. You're looking for ‑ but I mean pharmaceuticals are not ‑ they are designed to be not toxic. They're designed to be good, at least most of them, and so you're going to have to go really, really high in concentrations to find toxic effects of pharmaceuticals. Instead, you will find therapeutic effects or pharmacological effects at very low concentrations, but those effects will usually go unseen in these ecotox tests we're doing today and I'll come back to that at the end of my talk.

They are also persistent or semi-persistent in the environment. Many are designed to be really stable because the human body is a hostile environment and it has to stay as stable as possible because many of the metabolites of pharmaceuticals can be dangerous for us or have even stronger effects than the other compounds.

Another thing that makes pharmaceuticals important is they are acting on drug targets that are often evolutionary conserved across phyla and this was shown beautifully in 2008 where they pretty much showed that when it comes to pharmaceutical exposure, a fish is very, very like a human. We are sharing a lot of these drug targets and hence, the effects of pharmaceuticals in humans might also be present in fish.

Finally, many of the pharmaceuticals have been shown to be bioaccumulated and bioconcentrated in organisms, and some have been shown to biomagnify, but it's not as all as straightforward as it is in many of the legacy contaminants where the biomagnification is really predictable. It's not at all like that in pharmaceuticals and we can talk about that after this talk if you want to.

The first time I got in contact with any effects of pharmaceuticals in the environment is this classic study or this classic effect, this tremendous tragedy of the vultures in South‑east Asia, where vultures died like flies, 99.7 per cent of this particular vulture was eradicated in just a few years. No‑one knew why. As it turned out after many years of detective work, they found out that it was Diclofenac, the anti‑inflammatory drug that they give to livestock that caused kidney failure in these vultures and they died. A side effect of a pharmaceutical in the environment that no one could predict before it happened. It had tremendous ecological effects because not only did the vultures die, they are the ones that are cleaning up from what livestock walking around freely and toppling over and dying. When the vultures weren't there, the wild dogs stepped in and increased exponentially with a huge burst of rabies following and also the wild dogs wrecking havoc in the natural environment, killing and reducing population of a lot of native fauna. This was the first time I got in contact or heard about ecological effects of pharmaceuticals in the environment.

Apart from this study, there's also another famous one that also showed ecological effects of pharmaceuticals in the environment, which is the intersectionality of wild populations of fish, mainly roach in the UK because of the contraceptive pills that were excreted and going out into the wastewater. This was a long time ‑ not a long time ‑ this was 20 years ago, little bit more even, that they found out and showed this, and we still can see about one fourth of the roach in the UK show signs of sex-reversal still, almost 25 years later. We haven't really dealt with the problem. I was at a CTAC meeting not long ago where we were discussing the level ‑ at what level should we say that it is okay to flush out ‑ at what level can we accept contraceptives or hormone disruptor compounds in rivers and streams? We're suggesting that no effect concentration ‑ predicted no effect concentration, but then it turned out that over 75 per cent of all streams in the UK would not pass this level. So they had a suggestion that we double the level because then there would be ‑ wouldn't be a problem any more. The problem would have gone away. For me, as an ecologist, that is really not the way to solve a problem. The problem doesn't go away just because you increase the threshold that you can accept. The problem goes away if you remove or reduce the exposure for the wildlife.

This sparked my need and my will to improve the ecological relevance of ecotoxicological testing and also of the risk assessment of pharmaceuticals and other compounds.

Most studies of the pharmaceuticals have been done in the lab, and I started in the lab too. These are a few of the creative samples of lab studies that have been done looking at how pharmaceuticals can affect behaviours or other ‑ I would say behaviours connected to recollection, and these, of course, are extremely important behaviours, but also behaviours that are usually very hard to change because they are really important for the organism and there has been selected - with strong selection over a long time. The fact that these pharmaceuticals can change these behaviours that are so strongly selected for is really showing that pharmaceuticals are a powerful contaminant that can be very dangerous for the environment at high concentrations. At rather low concentrations, but high enough to cause harm.

I started by saying that I was an evolutionary and behavioural ecologist, so my first thought of increasing the realism of pharmaceutical risk assessment or ‑ yeah, increasing the ecological relevance of ecotox tests was to use behavioural end points for behaviourally modifying drugs. That's sort of a no‑brainer because behaviour is really the link between the physiology of the individual and its environment, and behaviour is a very, very sensitive end point as well because before you start to have individuals dying, they sure will have changed their behaviour and it's been shown that many times behavioural end points can be up to 1,000 times more sensitive than their regular common use end points such as morphological changes or reproductive changes or … end points.

Then when I started back in that cold room without no windows, end points had received fairly little attention. There's been some behavioural studies, but most of them had been on looking at behaviour as a side effect, but they took notes that fish were swimming upside down in the tank while they were exposed to this chemical, which is not really the behaviour I was looking for. I was more interested in looking at behavioural end points that were ecologically relevant and had a solid ecological theory that I could back it up with.

In that room, we decided that we wanted to use an ecologically relevant drug with ecologically relevant concentration on an ecologically relevant species and look at ecologically relevant end points in that species. The chemist I never met before suggested we should use one of the benzodiazepines. He said I will go and check which one is most prominent in the rivers we've been sampling. We go out to sample more and then we'll decide. When he came back, he suggested Oxazepam, which Minna found in her water sampling as well, because it's a very commonly used benzodiazepine. Benzodiazepines are anti-anxiety drugs that are also used for inducing sleep in elders, for example. There are about 20, 25 of them. Many have been taken off the shelf, but some of them are only sold on the black market because they are narcotics. Many are used and many are excreted in the same streams.

Oxazepam is one of the least potent of the benzodiazepines. I want to say that before I move on. They work on the GABA receptors, subunit A. This is the evolutionary concerned drug target that benzodiazepines work on. It's both in humans and in all other vertebrates. We expected that if it changes the behaviour in humans, it might also change the behaviour in fish.

Minna has already talked about how pharmaceuticals end up in the environment, but where in the environment do they end up? We went out, like I said, and sampling a bunch of rivers looking at benzodiazepines, where are they, what kind of benzodiazepines can we find? Are they even there? We started out in smaller rivers with a lot of people around. This is an example from the UK, the area called the river basin and as you can see, we found a lot of benzodiazepines in small rivers. Then we upped our game and looked at the huge river, Danube, that runs through most of Europe. A really big river, a lot of dilution and we find benzodiazepines all along the Danube which was both exciting and disturbing for us. Since then, we moved out into the ocean to test if we can find pharmaceuticals even in the sea. To the right, you have the Baltic Sea. Sweden is to the left of the map, the blue sea, and Finland, Minna, is to the right. The numbers on that sea, 1 to 43, I think, are the sites where we took water samples at surface water and 20 metres deep. In all these sites, we find pharmaceuticals, except 1 and 2, up north where no‑one lives and there's a huge input of fresh water because they are large rivers coming out from uninhabited areas. From sub site 3 and down, we find pharmaceuticals in all our samples, and many pharmaceuticals when we get close to the land and fewer out in the Atlantic areas.

One pharmaceutical we found in all samples in the Baltic is Carbamazepine which also Minna found. It's a super stable drug and we had an engineer calculate how much Carbamazepine there is in the Baltic Sea and it turns out that 89 per cent of all prescribed Carbamazepine around the Baltic are now in the Baltic. There are still no half life of Carbamazepine in the wild because it's so stable. So we're slowly marinating our fish in the Baltic. I'm doing a collaborative study with a Swedish marine agency, water and marine agency about looking at the effects of pharmaceuticals in cod because we've seen high levels of hormone disrupted compounds in cod in the Baltic Sea, which is disturbing. 5,000 times higher than what shows reproductive change in lab, which is ‑ that's just mind blowing. We're also sampling outside of Florida together with Florida International University looking at the potential effects of pharmaceuticals on the collapse of the bonefish population there. That's a famous and liked sport fish.

So basically wherever we looked we found pharmaceuticals, but admittedly most of them are in very low levels in the ocean, in the sea. But they are there still and they are increasing.

So now, over to what we are doing. It's a long background on my way into this spiel, but now in the lab we are looking at ecological effects of human activities and pharmaceuticals, climate change, all these together. We're both looking at the pharmaceutical effects singly, alone, but we're also looking at indirect effects of other things at the same time. So how, for example, the interaction of climate change and pharmaceutical will act, because are we really measuring the risk or assessing the risk of pharmaceuticals for the future or is it just valid for today? If we have a temperature increase of 2 degrees in the water or the maximum degree temperature increase in the water of 5 degrees even, what will that ‑ how will that affect the risk of being exposed to different chemicals and pharmaceuticals?

So we're looking at these direct effects and indirect effects of human activities basically, and how this effect, predominantly fish, but also insects and crayfish and all other things. In the lab where we look at mechanistic, realistic ‑ or mechanistic and low complexity things, why ‑ what happens? What's the reason why we see this? Then we move into artificial ponds where we can follow the fish and see if we still see the same effects in a semi-natural environment, and finally we have a lake area in the interior of Sweden where we can do full lake studies following individual fish on high resolution, and I'll come back to how we do that later. But then we can actually get to the $100 million question of what is going on in the lake with all the complexity and interacting effects that's out there.

Back to that room. No windows, cold, me and a chemist. The most important thing that we did before going into that room was leaving our egos outside the door. We said that, okay, let's just be ... just talk openly, no prestige. That was very important because we were all shot down by each other at least 10 times before we reached the final design of the study, and the design was, as it turned out, really successful. I have the chemist to thank for that because he chose the compound and that's the most important thing in this case. But we chose the ecologically relevant species, perch, because it's a really common species all over Europe and Asia. I think it's an invasive species in Australia, or at least it's introduced. We decided to expose it to realistic levels of the drug Oxazepam that we found in the river samples that we looked at. What we saw of the behavioural end points we looked at, because we decided to look at, since I've been working with animal personality, there is a big five in fish or animals as there is in humans - activity, exploration, sociality, boldness and … anyway, we chose to look at activity, sociality and boldness - boldness is risk taking, basically. What we saw when we exposed them to these realistic levels is that it increased their activity and it reduced their sociality. They were more active, less interested in schooling ‑ this is a schooling fish ‑ and they also became, especially at higher levels, more bold, more risk taking. We did this experiment and we realised that, okay, now we know that they change behaviour, but what does it mean? How does it affect the perch's ability to be a perch? Then we redid the entire experiment once again and we added a feeding ... to see how it would affect an ecological end point, how the behavioural changes would affect an ecological end point like feeding. That is very important for fitness and growth.

Again, we saw the same behavioural effects and to our surprise we saw increased feeding efficiency in the perch. So what the drug actually did was making the perch better. They fed faster, they would grow faster with this anti‑anxiety drug Oxazepam in the water. But then of course we did ‑ there is also a flip side of that increased activity, reduced sociality, increased boldness. If you are more active and less social, you don't seek protection with the group and you're also more risk taking, you're going to expose yourself more to predation. We did a predation trial as well and saw that while we had actually pretty awesome ‑ awful, I mean, to look at the videos, because we had a staged huge tank, many huge tanks, 3 x 1 x 1 metre, one pike which is a predatory fish, two perch, one exposed and one non-exposed in the other, then we open up for the fish to interact and video recorded all these interactions. We did that 18 times. 16 of the 18 times, the exposed perch was eaten first by the pike. The other two times, no perch was eaten. So the normal unexposed perch were never taken by the pike in the presence of an exposed perch. You could also just see on the perch when you introduced it into the tank, even though it could not see the predator, it felt the chemical cues of it, because the normal perch turned black and just sank to the bottom and lay there, I am a stone, I am a stone, I am a stone, whereas the exposed perch were just swimming around happily as ever. Some of the videos even show when you open up the doors for them to gain access to the entire tank, some of the exposed perch seemed to just see the pike on the other end and swam straight up for it and wanted to know what's going on over there. So they sort of removed the fear in the perch which was a deadly consequence, which of course in the field would also be very detrimental for the perch population.

Luckily, another study showed that pike also becomes sloppy feeders when you expose them to Oxazepam. That's a different story.

Okay. Then we knew what's going on in the lab and we wanted to know, okay, is this true in the field as well? So what we did here was we ‑ this was our first attempt to bridge the gap between lab and field and here we did lab exposure of fish and then we introduced them in a fishless lake, in this lake area that we have, and the blue dots are the predatory pike and the red dots are exposed perch and the yellow dots are the unexposed perch.

Now I'm going to see if I can get the video going. No, I can't. Maybe it's the remote control that is screwing up the video. Anyway, there it is. Perfect. No? Little help from our friends. Thank you.

So the blue dots are the pike and we introduced them first, just so they’d get familiar with the area, then we introduced the perch, and the red dots are the exposed ones. You will see the red dots starting to move out while the yellow dots are still in the introduction area. If you look at the lines being created in these squares, you have a red line and a yellow line, and the red line, of course, is the exposed fish and the yellow line is the non‑exposed. What these curves show is that fish that were exposed to Oxazepam were more risk‑taking than fish that were not, and fish that were more ‑ were exposed to Oxazepam were also more asocial. If you look at the lower right graph, it's distance to nearest two neighbours. Red means they were more asocial. We've shown the same behavioural effects in this short‑term field study. This is just ‑ you can see it's five, six days, because if you see in the upper right, it's the elimination curve. If you take fish that has been exposed to pharmaceuticals and put them in clean water, it slowly enters steady state with the water. They clean themselves all the time.

This was exciting and scary. We've shown that what we see in the lab can also happen in a full lake, in the field, the behavioural effects.

Let's see if I can ‑ did I…

That's the first example where we took a step out into nature and tried to bridge this gap. This is the start of the second attempt. I'm working in a really salmon heavy department at the moment and it was expected of me to work with salmon as well when they hired me, so I thought, okay, let's see if pharmaceuticals might affect something really, really important in salmon, their migration. So we did a lab study and these are migration chambers, tanks you can see here in the picture. They are eight metres across and are fed with river water, and you can measure ‑ if you mark the fish with tags, which is basically barcodes that you can use to identify fish, and then you measure how many laps they swim in a certain time. They should swim downstream because these are small salmon moving downstream from their river to the sea. What we saw when we looked at exposed and non‑exposed was that exposed fish were migrating much more efficiently or with high intensity. They would reach the sea faster than the non‑exposed, and the guys at the aquacultural facility were really excited about this. They were like, wow, now we need to prep all the salmon with Oxazepam because the most dangerous part of a salmon's life is when they're moving from a familiar habitat downstream through an unfamiliar habitat with all kinds of dangers into the sea. So they wanted to reduce the time that the salmon was exposed to these dangers.

Now we said, hold on a bit. Not yet. We want to test this in the field as well. Is it really valid? So we went out into a small tributary, and we put out this ... again and looked at how long did it take for these salmon and trout later as well to migrate from our release point down from the mouth of the tributary into the river. Again, we saw the same effect. They were faster, much faster when they were exposed than when they were not.

The same goes for sea trout, also migrating species that migrates out to the sea. In this case, we used Oxazepam for the salmon and Temazepam, another benzodiazepine for the trout, we saw the same effect. The guys at the aquacultural facility were ecstatic. Yes! Now we've shown it. Let's order some Oxazepam and expose them so we will have ... survival. Not yet. Not yet, because we need to test it in the real river. The Umer river with all its complexity and all its predators and everything that's going on there, one final study. They said okay. Let's see. These scientists. Then we used acoustic telemetry as a method to track the fish instead of tagging them. Acoustic telemetry is a nice and fastly developing field of field equipment that you tag a fish with a small transmitter that sends out an individual signal that makes you able to identify the individual fish, then you have receivers that you put out along the river or in a grid if it's in the lake.

So we did this in the Umer river and we exposed fish to Oxazepam or had control fish. Then they migrated from a position upstream ‑ it was 27 kilometres upstream ‑ and all the way down to the gulf. When we got the results from this study, we were less keen on exposing them. This is real ugly ‑ it is what it is at the moment. We had 18 salmon for treatment, 18 exposed and 18 unexposed. Even 100 metres after we released them, we already lost two of the exposed fish. They never reached their first receiver. Half the way through the migration we already ‑ we only had six individuals left out of 18 of the exposed ones, whereas we had 15 of the normal fish. Only three of the exposed ones made it to the ocean or the sea, whereas 12 of the non‑exposed.

Here we've shown that even though it looks like ‑ and the ecological theory predicted that Oxazepam should increase or ... probably should increase the survival of the salmon because it would reduce the time it spends in the river, it did actually the opposite. It reduced survival and as it turned out, we had a great French intern that worked really hard trying to find these transmitters in the river where they were, and also he was an extremely keen fisherman so he fished a lot. He showed a beautiful correlation between where the transmitters ended up and pike densities. These fish that are exposed to benzodiazepines get eaten by pike and perch in the river. They are not afraid, swimming fast and don't care about risk at all. So no exposing of salmon to benzodiazepines in that aqua-cultural facility at least.

Okay. Moving on. Now I'm almost out of time, so I'm going to end my talk with a couple of other things that we are doing that I think is really exciting and important. One is this study that I did collaboration with Monash University and Erin Richmond, Bob Long and many others. I was going to Monash ... Minna way back and stumbled over this fantastic material that was just lying around. Me and my pet chemist were excited about it and brought it back with us to Sweden and analysed it and from that, we were able to show that pharmaceuticals actually are flowing from the streams up into the terrestrial environment via aquatic emerging insects into spiders and other animals like bats and birds. There are more studies on this topic now. This was the first one when it came out, but now it's been shown repeatedly and we also did a huge study in Sweden where we had 600 sites sampling water and invertebrates and spiders. Unfortunately, with these huge studies, it takes a long, long time to publish them because there's so much data. It's still not published, but it will be. But this is something really important and new because we found really high levels of pharmaceuticals in the riparian spiders, much higher than in many of the aquatic insects. So not studied enough field, this pharmaceutical flow.

Another important thing that we're doing now is, of course, looking at individual compounds versus mixtures, and not only mixtures of mother pharmaceuticals or real pharmaceuticals, but also the effects of exposure to the metabolites that the pharmaceuticals turn into. We had a publication recently where we showed that fish can actually be more ‑ we exposed the fish to Oxazepam and another benzodiazepine which break down - the metabolite of that benzodiazepine is Oxazepam. The fish were more exposed by the breakdown of the other benzo to Oxazepam and to Oxazepam itself, which is fantastic, so the fish that were swimming in one of the ‑ for example, in Temazepam, were exposed to more Oxazepam than the fish that were swimming in Oxazepam. So that is another thing that ‑ another issue that I think is not studied enough, the combinatory effects of not only the individual compounds, but also the metabolites and how that in turn might spiral into increased bioconcentration and increased ... effects.

Finally, I'm going to end with wild-caught zebrafish versus lab zebrafish. This is something that's been a pet peeve of mine since the beginning. I think at least when we're looking at behavioural end points, we can't use laboratory species, or species that have been in laboratory for many generations because they've been selected for a certain repertoire of behavioural characteristics. They often live in a luxury environment with no predation and no risks. We did an experiment together with … Vossen where she looked at wild-caught zebrafish straight from the Bengal and laboratory zebrafish that have been bred in the lab for hundreds of generations and exposed them to Oxazepam for seven days and in the wild-caught zebrafish, we found clear behavioural effects and in the laboratory zebrafish, absolutely no effects at all. This is disturbing since when we look at doing risk assessments, we often use laboratory‑bred zebrafish or species like that, but they might not ‑ at least when you look at behavioural end points, represent the real risk that is to wild animals.

Okay. With that, I want to acknowledge all of the fantastic people that have been working with me in the lab. These guys have done all the work. I've been looking and nodding and humming. I want to thank you all for logging in, listening and I have to thank the money, of course, that made all this research possible, and I also want to make a little push for ‑ there's a paper that came out a while ago about the role of behavioural toxicology in environmental protection and please have a look at this paper. I think it might be interesting for many of you. With that, I end my talk and I'm happy to try and answer any questions. Thank you.

- Tomas, that was just mind bogglingly frightening to be frank. It reminded me of my first exposure, so to speak, to this issue which was in a movie called The Disappearing Male and I don't know if you've seen that movie, but that was over a decade ago I saw that and it really sparked my interest in the topic. Your work also reminds me of Rachel Carson's journey many years ago and it's an emerging problem and what I can see from your talk is the consequences of these pharmaceuticals, not only on the animals, but on the outcomes for the animals, and it's made me think, when you were giving this information, if it's doing this to the fish, it must be doing it to the people, and does it mean that the people who are using some of these pharmaceuticals, for example ‑ and I'm thinking about the Oxazepam, it's the same for many of the drugs, whether those who are involved in road accidents are people who have been exposed or have greater exposure because they take greater risks. You must have thought ‑ no doubt, you've thought of that during your many journeys, and somebody will be picking that up or looking at it. There will be some great opportunities, and of course the final comment you made in regard to the errors that we might ‑ systematic errors we might be making in our toxicological experiments using lab versus wild fish, I think that is really deep insight and we could be making gross mistakes about understanding the effects of these pharmaceuticals, not only on the species, but on humans.

You may wish to comment on some of that and then I can move into some ‑ a myriad of questions that I've been provided.

- Yes. First of all the human aspect, and I've been so tempted to do some human behavioural tests and studies, but it's really hard and I've been staying out of it because of the extreme hassle of doing human studies ethical‑wise and also the problem of getting the information of who is actually using what product. It's very ...

- Putting my professor hat on, Tomas, I make a suggestion to you. You could really go and look at cadavers because you would understand what happened and that may be a way to start some of that preliminary research because the cadavers have given their bodies away for scientific research. That would be possibly a good project for an early researcher, an Honours project or a Masters project to look at that. Anyway, sorry, carry on.

- Yes, and for the second part, I truly think that we are ‑ at least for some compounds, gravely underestimating the ecological risk. The assessment is not designed ‑ the testing itself is not designed for testing for risk of pharmaceuticals. It's not testing the therapeutic effect and we are not using the proper, in some cases, the proper organisms or the …

- No, it's pretty interesting. If I can go to some of the questions and answers that have been provided by the audience, I've got five questions, it says here. Oliver James has raised the question, this is fascinating work, which I agree with, and great talks, including Minna. They're asking of you, Minna, are you going to publish the data? If so, where? As a journal article or a report?

- Good question. Thank you, Oliver. Yes, we are publishing as a journal article and this is good knowing when we have such an audience to share the news that EPA does publish data as journal articles as well, maybe traditionally we've been publishing them more like technical reports or EPA reports, but there is definitely more emphasis now on writing journal articles because they are more openly available for a broader audience and you can actually see ‑ also the peer review is important. It's important that any work that we do goes through the peer review and then we get the feedback and then we can improve work like any scientist working, acting in our government. Yes, journal article is coming up. We are writing it as I speak.

- That's fantastic. As the Chief Environmental Scientist, I will not encourage you enough to do that and I and my office will be there to support you. Related question from Dr Amy Heffernan was: how many PPCP did you test for? Was it 72?

- It was 72. 72 in the water and 22 in the fish.

- Thank you very much. Next question on my list - Tomas, you can relax or ignore me.

- Fantastic.

- The questions are coming through for Minna. I have received a few for you so I will come to you in a minute. Minna, an anonymous person is asking: what were the climate conditions when the samples were collected? Dry times, discharges, etc? It's a question that also crossed my mind. Was discharge an influencing factor on the concentrations that you measured?

- Very good point. Yes, as I mentioned, we started sampling in March, which in Australia it's late summer and early ‑ went from March to April. It was around the first lockdown. So there were some delays because we were not able to do everything within four weeks as planned so there was a window of two ‑ six to eight weeks that we were going ‑ collecting the samples. So there would have been different types of weather conditions, but yes, it's very important to consider all those factors when we start interpreting the results. What do these mean? Because definitely the water conditions, and also as I mentioned the flow of the waste. If there's not that much waste coming through the plant, then the levels might be lower and therefore then whatever the outcomes were, how we estimate the risks might be biased because of the conditions that were present when sampling. We had effluent samples as well, but we didn't have it for all the sites, that's why I didn't present that today.

- I will leave it with you to think. The same person asked about counting for fish mobility when it assesses their exposure. I want to move on to Tomas to make sure we have enough time for questions. Question from Paul asks to what extent do some of the pharmaceuticals in water bioaccumulate up the food chain? For example, are there more fat soluble drugs in big predatory fish versus small prey fish?

- That's a good question and it's a complicated question because it's so compound specific and species specific as well. … content does not explain more than ‑ less than 50 per cent of the variation in bioaccumulation in fish, bioconcentration. There are other things going on with pharmaceuticals than regular contaminants. We haven't seen ‑ in our studies, we haven't seen a straightforward increase in every step in the food chain, rather it depends on the ‑ more on the physiology and the exposure routes of the organisms. We also see really, really high levels of many different pharmaceuticals in snails, for example. So anything eating snails are in big trouble because ‑ probably because a snail is really exposed to the environment, to the water. So they bioaccumulate a lot. Then they also eat algae and algae are also very high in pharmaceutical content, the ones that we've been looking at at least.

When you go up into fish, they ... against the water. They are often lower. They often are exposed to lower levels than the snails and the algae, because they have these active getting rid of excretion system that they use. So they sort of end up in a steady state with the water. Bioconcentration can be really ‑ I mean vary a lot between species. Even within fish it can vary between 0.8, for example, with the crucian carp that has less pharmaceutical in their bodies than what we found in water. This is for Oxazepam, whereas perch that we started out with, looking at the effect on perch, they have about 12 times higher concentration in their bodies than there are in the water. This means that if crucian carp and the perch swims in the same pond, the perch will be exposed to 14 times higher concentrations of Oxazepam than the crucian carp even though they are in the same water, which means that many of the pharmaceuticals will have asymmetrical effects in food webs and asymmetrical effects, that means they affect some individuals or some species and not the other. The asymmetrical effects are the ones that we are most afraid of because they are most likely to have huge consequences for the food web and for the ecosystem.

- What you're saying reminds me somewhat of emerging understanding of PFAS and we know that PFAS accumulates and is removed from different animals at different rates and we typically rely on animal models to try and understand human health effects. I think what you're saying ‑ probably what you're saying if I can unpick that, is we can't probably rely on animal models because they all behave differently in response to pharmaceuticals which creates a very significant problem in understanding what constitutes a safe level and what the dose response is, and somebody has asked also, do any of these pharmaceuticals ‑ is there an established threshold of safety?

- There are a lot of thresholds of safety and predicted no effect concentrations of all of them, but they are all based on the regular ecotox tests that you do for any chemical.

- Lab fish?

- Many of them ‑ I think that most of them are way too high.

- So actually, you've just opened the door on unpicking that. That would be a great piece of research. You wouldn't have to get your hands dirty on humans, you would be able to demonstrate that the toxicological assessments are invalid. That would be a really insightful piece of research and you've probably already thought that anyway, so I don't know why I've mentioned it. Moving on to other questions, this is a question for Minna. What risk assessment were you looking at? Was it a single compound or a mixture of compounds? For example, what is the risk when you consider all of the drugs together and Tomas, you may also want to chip into this ‑ including other contaminants such as PFAS and other endocrine disruptors that the fish may be exposed to and how would that relate to human health risks is the other question that comes on there.

- That's a good point and actually links to what Tomas was saying. The risks assessment that I provided was based on single chemicals, and therefore even though based on single chemical analysis the risk is low, the fact is that the animals are exposed to ‑ when you eat a fish you are likely to be exposed to a mixture of chemicals that are in the fish, that we only measured 22, there have got to be more than 22 in there that we didn't even look at. The risk assessment should be based on mixtures, but we haven't done that yet. That work, and as Tomas mentioned, it is quite a new ‑ there's a lot of risks assessments out there and the guidelines are limited, so it's a bit early to talk about safe levels. More work needs to be done, but what are you thinking, Tomas? What would be your advice for the regulator?

- It's complicated, as always. One of the things that strikes me when you look at mixtures is that there are many, many, many mixtures or combinations of pharmaceuticals that would never, ever be prescribed to a human being at the same time. Completely forbidden. Because they have such huge side effects to the other, but these fish are exposed to the combination of many of these dangerous or bad combinations that would never be prescribed to humans. I haven't seen a single study looking at those types of mixtures. You’d rather look at mixtures of different types of anti‑depressants or different types of ... but what about those types of mixtures? I think that would be a really interesting avenue to open up.

- Tomas, I would wager that you would not be able to find a safe threshold for most of those mixtures. I think that's what the outcome is for the assessment of many chemicals and compounds. The more we understand about low level exposures, often the exposure is greatest per unit of exposure, the effects are greater per unit of exposure because it's a super linear curve, the first and lowest exposures. I suspect we're only at the beginning of that journey and you're lucky to be at the start of that journey. I have got some other questions that have come through that I would like to ask.

Maybe we can talk about what could the community do as individuals to contribute to reducing the pharmaceutical load going into waste waters? It's a fairly straightforward question, but could you cast your professional mind to that and give some advice?

- Should I start?

- Yes, Tomas, and then, Minna.

- Okay. First of all, ask your physician or whoever prescribes your medication if it's the environmentally friendly version. There are many options when it comes to most of the pharmaceuticals that you can take. At the moment, there are very few that are green drugs. There is a movement towards having a stamp, sort of a green fish or something like that on some of the pharmaceuticals so you are able to choose the better product for the environment. But we need to ask for it first. So ask for it, because that builds the pressure for the pharmaceutical companies to develop it. Otherwise, it will take a long time. That's one proactive suggestion.

- Thank you.

- Apart from that, I don't think anyone should reduce their medication that they need. We need ‑ many of us need medication and might not be around without it. I don't want to advocate people stop using pharmaceuticals at all, but I think that the solution ‑ the short‑term solution ‑ rather short‑term solution ‑ is cleaning of the wastewater. That's where the rather short‑term solution is, and the long‑term solution is green drug design. But both of those are not things that you can do at home. So don't flush the pharmaceuticals down the toilet. Return them to your pharmacy and ask for green drugs.

- Excellent. Minna, do you have anything to add to that?

- Yeah, just a minor thing, speaking to the Australian audience, there is a return unwanted medicines scheme in Australia. So if you Google that, you are able to find the pharmacies that actually take your old or unused medications. So it's worthwhile checking because not every pharmacist will do that. So when you're moving out or cleaning now that we're in lockdown, I'm sure we are cleaning our cupboards, so pile those up and then check the website and take them to a pharmacist that will look after them properly.

- Thank you. So a couple more questions are coming through, and just give me the nod when you decide you would like to terminate, you've had enough and you need to get breakfast or something, Tomas, but one of the questions is: how often should we check for these pharmaceuticals in water supplies and samples? One question that's in my mind related to that is if we think about our Murray‑Darling system, we have multiple wastewater treatment systems disposing water upstream of a village or a town where they then take the drinking water and we know that standard water treating process typically does not attenuate or remove the pharmaceutical products. It probably accumulates, I would estimate, as you go downstream. So what should we do? Should we be adding the testing of these chemicals to our water supplies and also in testing of our field samples? What's the sort of frequency you think we should do and should we do this sort of work?

- You can start, Minna.

- I can start and I'm sure we ‑ please do add in, because again there's no one way of doing this, but based on what I know that I've been now part of designing the first kind of sampling design, I would say, let's start at least doing ‑ look at seasons, difference between seasons would be my starting point and use the passive sampler … then based on those results, then use that to narrow down do we need to do it that often, but also let's visit the same sites multiple times because without ‑ when you do a single study like we did, yes, we have a snapshot, but it's not going to tell us about the average concentrations that are out there. But then again there's ‑ with this one you can go ‑ you could… the more you do, the better understanding you get. There's almost no limit to this, but I know that there needs to be then ‑ you need to draw the line somewhere. I would say at least try every season and then do it in a targeted matter, but when you do sample, sample well. Sample different biota, water and more than just fish. We didn't look ‑ as I said, no spiders, no invertebrates, nothing. So there is definitely when you do sample then, collect as many as you can. Tomas, any other ideas?

- No, I completely agree, especially ‑ I mean it's really important to sample biota because to be honest, it's not really what's in the water that is important, it's what's in the organisms, because it has to go into them to have an effect. Some compounds you can't measure in the ‑ or you won't find in the water, but you will find it in the organisms because they bioconcentrate it and it's below detection level in the water, but it's high enough in the fish, especially ... compounds.

If you have such a situation, I would sample the drinking water in that village because I bet you can find 15 pharmaceuticals in the drinking water.

- People might be a bit concerned about actually doing the sampling for fear of finding something, I suspect. Just related to this, two questions. One is: will EPA be continuing the monitoring to develop good time series and related to that, as a result of the ... that you've acquired that showed there was elevated levels upstream from the discharge point, they've also asked: how do you explain those results?

- First question was: will EPA continue this? At the moment, we don't have ‑ so the funding for that was given just to do that one sampling ‑ that sampling program that I presented. In the near future, we don't have resources for that. Hopefully when we publish this in the near future, there will be more support for this work. At the moment, no, this was a bit of a one‑off study without follow‑up work. However, we are working ‑ we're analysing recycled water. We have a project on looking at recycled water and that is funded by DELP, and that's work that the sampling has been done for that and we've been actually analysing influent and effluent waters. This time we do look at environmental samples, we went back almost to the source looking at influent and effluent, what comes in and what is treated and we've been analysing those samples for pharmaceuticals, so personal care products and EDCs and we sampled 30 different SDPs across Victoria and sampling was done between April and June this year. So that is a work that ‑ we will report that in the next 12 months.

- I thought all water was recycled, but I'm just being a bit obtuse there. Which it is true. Another question is a quite important question about dissemination of knowledge: what types of publications or educational materials are available to be distributed to pharmaceutical waste generators, or better still, to inform them of the long-term environmental risks associated with waste residues being discarded down unfiltered drainage?

- Very good point. Yes, I would say probably quite limited still and quite an emerging area from the ... point of view and my understanding is when you look at different states as well, Victoria's leading the pharmaceutical work at the moment so there is probably ‑ when you look at the website there's quite limited … so we'll definitely ‑ that is something that we can do and will be doing, providing more information on the website, but then hopefully, Tomas, I'm sure some of your articles will be on open access, so actually public can access the scientific research to some extent so the information is there, but it doesn't mean that it's always, I understand, in a format, language that is easy for anyone to read, but there is information, but there's definitely room for improvement in terms of giving guidance on how to ‑ what can ... do.

- Tomas, what do they do in Sweden about this problem?

- Thankfully they are ‑ it's started to move. The EPA in Sweden has devoted $250 million per year, starting three years ago for five years, consecutive years, for improving or implementing pharmaceutical removal techniques in wastewater treatment plants in Sweden. So we still have one more year to ‑ I'm a part of the evaluation group of those applications, that wastewater treatment plants can apply for money from the EPA. 90 per cent of the cost of the technique implementation. Then depending on how much environmental benefit it will give, we evaluate the applications.

So that's a great step forward at least that we are sort of trying to find the worst spots and remedy those. Apart from that, we have a centre for pharmaceuticals in the environment that were started two years ago ... the Swedish medical agency that runs it, and it coordinates all Swedish researchers and decision‑makers and the industry and tries to be sort of a spider in the net, finding and increasing the information flow and the collaboration between the different groups. Also, regulators are an important part there. Apart from that, I think it's ‑ we're starting to implement it on the county level as well, the pharmaceutical sampling in different ‑ like a monitoring thing in different streams. So the ... have increased enormously the last 10 years, I would say from nothing to what it is today. It's really ‑ I'm really happy about it.

- Yeah, that's fantastic. It seems, as with PFAS as well, the Nordic countries are ahead of the curve. In relation to understanding the toxicological effects, Paul has asked a question. He says: where we can't measure the effects using traditional end points such as liver damage, what about using other molecular biological techniques such as ... for example, or DNA methylation, those sorts of modern techniques. Are you involved in that or do you have colleagues?

- I am involved in that a little bit. I am co‑supervising a PhD student, that is ‑ together with a chemist and she's doing metabolomics, looking at how waste water exposure affects metabolic profiling … she’s using ... so an invertebrate as her study organism. She's published two papers so far on this topic and has a couple more coming, but metabolomics are tricky because it's ‑ you see that things happen, but you ‑ it's really hard to know what, especially when you're looking at an inspect where many of these pathways are just ‑ no‑one knows. So it's not ‑ you can't sort of translate the pathways that they represent in humans to what they do in insects. It's definitely an exciting and interesting tool. I encourage everyone to try and figure it out because that would help a lot. I'm excited about pursuing this as well, especially trying to identify pathways that might represent behavioural effects and other effects that are really hard to see in the field, or in the lab as well.

- Tomas, Minna, I think the questions are slowing and I think we can draw this to a conclusion. I would like to say, unless you have anything else to say ‑ do you have anything else you want to add today? No? Thank you very much for your time, both of you, preparing your slides and giving your afternoon up to tell us about this really important and fascinating research. It looks like we're really at the beginning of a long journey where we're beginning to understand the consequences of treatments that have been applied to people which then filter into the environment, and it really does look like a Rachel Carson moment in time. So I would also like to thank everybody who has attended this livestream seminar today and for the excellent and interesting questions that have been submitted. We look forward to seeing everyone at our next environmental science series event which is in October 2021. The topic of which will be microplastics. Tomas, Minna, everybody else, thank you very much and good afternoon and good morning to you in Sweden.

- Thank you very much.

- Thank you.

- Thank you, bye bye.