What climate change does to kelp forests
In this episode, how climate change impacts kelp forests, selecting for less animal-friendly variants, refining AI models for better water infrastructure design, classifying extinct marine megafauna and when best to swim with them, the coast consequences of climate change, and why a better understanding of the planet's drylands is critical...
In this episode
00:54 - Climate change impacts kelp composition
Climate change impacts kelp composition
Sarina Niedzwiedz, University of Bremen
But first, marine ecosystems are built around plant and algal species that capture solar energy and feed it up the food chain. Kelps are major players, especially around the Arctic, and, as well as being a source of food, kelp forests also provide shelter and breeding grounds for marine life. But a new study suggests that some species are very vulnerable to climate change, which can alter water temperatures and water turbidity, which can affect the amount of light that makes it through. These altered conditions exert a selection pressure, shifting the balance of kelp species away from beneficial forms towards those that afford animals a less hospitable environment, which can have economic costs for us too, as Chris Smith hears from Sarina Niedzwiedz…
Sarina - My name is Sarina Niedzwiedz and I'm a postdoc at the University of Bremen and I'm a marine biologist. Kelps can actually be compared to trees on land because they can grow several metres in height and thereby build complex three-dimensional structures above the seafloor. They provide a lot of habitat and nursery ground for a variety of species and also food as their primary producers. So, they are very, very important for those ecosystems and act as foundation species.
Chris - And why do we feel that they might be threatened or impacted at least by climate change?
Sarina - Because the Arctic is one of the regions that is actually most affected by climate change and we see temperature rises at a rate about four times the global average and that leads to a variety of cascading other changes such as an increase of marine heat waves - but also all of this temperature rises cause glacier snow and permafrost to melt, and the huge amount of melt water that results from this melt is washing a lot of sediments into the fjords which is then changing the light availability underwater.
Chris - So, it's effectively a one-two punch that might hit the kelp then. Not only is the water temperature departing away from where they've evolved to be most happy but the water could become very, very cloudy and cut off the light supply.
Sarina - Exactly. Both temperature and light are very, very important drivers for those kelps. We need to know how the kelp forest dynamics is changing in the future.
Chris - And how can you get at this? Is there an easy way, because you can't speed up time and ask, well, this is what it was like before climate change, here it is afterwards. So, how can you get at this to get a reasonable answer to what the impact might be?
Sarina - We designed a multifactorial experiment where we exposed those two locally abundant kelp species to, first of all, different light conditions. Those light conditions mimicked either a clear water column or a turbid water column with all the melt water. And then on top of that, when they were acclimated to those light conditions, we exposed them to a marine heat wave.
Chris - And how do you gauge what impact that's having on the kelp?
Sarina - Yes. So, we measured their physiology, for example, their growth, their photosynthetic activity, and also their photosynthetic rates and respiration rates. And, on top of that, we also measured their biochemical response. So, their pigment composition, for example.
Chris - And when you do this, I presume you're looking at two different kelps, because then you can ask, well, what's likely to happen to a range of populations? But when you do this, how do those two variables: the temperature and how cloudy the water is, affect the outcomes for the kelps?
Sarina - When both of the kelp species were exposed to high light conditions, so basically a clear water column, they experienced physiological stress levels, which was actually destructive for them. This stress level got even higher when the high light conditions interacted with cold temperatures. And stress level was then mitigated by the marine heat wave. So, they actually benefited from warmer temperatures because both of the kelp species we were investigating are not Arctic endemic. So, their optimum growth temperature is warmer than that, what we currently see in the Arctic.
Chris - And of course, the warm temperature would also bring with it more meltwater, which would cloud the water, which would reduce the light, which would reduce the light stress.
Sarina - Yes. We also saw that with lower light intensities, the light stress got reduced, but the productivity and growth of both kelp species also got reduced.
Chris - Well, what do we take away from this then? It sounds like it's not necessarily all bad news for kelp, because there's going to be some that may do quite well from climate change then. They will be less stressed and they will retain their productivity. Is that good news then, that we can relax on that front?
Sarina - Physiologically, kelps can cope with those conditions that we will expect in Arctic near future. However, what we also saw is that depending on the conditions, the competition balance between both kelp species also changed. So, we saw that one kelp species, the sieve kelp, actually did better in darker and warmer conditions.
So, basically near future compared to the other species, the sugar kelp. The sieve kelp is also characterised by high contents of herbivore deterrents. So, if that kelp species, the sieve kelp, would become dominant in a future kelp forest, the nutritional value of the kelps might be reduced for higher trophic levels.
Chris - And hence, there would still potentially be knock-on effects further up the food chain.
Sarina - Yes, exactly. Up to the humans, because then commercially used fish or invertebrate species also are reduced in their abundance, then that might also have socio-economic impacts.
06:52 - AI aiding water infrastructure
AI aiding water infrastructure
Babak Zolghadr-Asli, University of Exeter and Queensland
The models that are increasingly being deployed to help water engineers keep the taps running. But, as Babak Zolghadr-Asli explains to Chris Smith, up till now, their design has largely focused on the quality of the model’s architecture rather than the quality of the data it’s processing. And, as with any AI model, when it comes to the quality of the output, the old adage “rubbish in, rubbish out” has never been more important. So making sure we have high quality datasets - and data collection methods - is essential to ensure the results and predictions are reliable…
Babak - It is far too easy to misuse these models or misinterpret their results. And imagine a case that you are embedding these models into some of your most vulnerable but most important infrastructures, like your water systems. If these models start producing results that are not necessarily reflective of the real world, then you have to face the consequences.Then you would face problems.
Chris - Is it the model itself, as in the way computationally it's working, that's wrong? Or is it the underpinning data, the information we are training it to use in engineering its own thought processes, that's at fault? Or is it both?
Babak - Very good question. There is this way of thinking that the algorithm is the most important thing. We are putting this in the centre and front and assuming every breakthrough is going to come from improving these models, becoming more efficient with how we are going to do about the computations and so on and so forth. However, there are those who believe that data is also important and should be at the focal point, because the model is only as good as the data that we provide it. So, the idea here in this specific case of water engineering being is that often at times we don't have enough reliable data to train these models. So, essentially they haven't seen the world enough so that they can replicate it in a good way.
Chris - Why is this not already something that people have confronted and dealt with when engineering these models?
Babak - I usually give this example.Imagine that you are a teacher and you want to teach your students. The challenge here is that the only exposure to the real work, your only material to teach this student is that you have a window to the real work and you're asking them to start drawing their real work through this window. Now imagine this window being really small.
Because of this limitation, the student is not being able to pick up every talent during this training session, right? Why haven't we seen any problem? The issue is that usually these models, for natural phenomena, it would take decades to see every pattern.
So for instance, we have to monitor and observe the performance of these models in these long decades to see if they are reliable or not. But we can't obviously wait this long because at the point that we realise we made a mistake, it could be too late. But the signs and the red flags are there because every now and then you would dig deeper in the performance of these models.
You have experts to go into how these models are trained. And if you go deeper, you would realise that the logic inside these models sometimes are not on par with the reality of the work.
Chris - What do you regard as the solution then? Is it just making these models more explainable? Is it having a sooner, more rapid feedback loop so that we can work out if they're going off kilter? Or is it just about enhancing the data that we feed into them so we don't settle for that very small window on the world that you use as a beautiful analogy, we go for the widest vista we can?
Babak - The issue is data. So, first things first, I think we have to manage our expectations and start seeing the problem for what it is. We shouldn't expect these problems to be fixed overnight as well. It would take years if not decades. But this sort of highlights the importance of starting investing in this. Because even if you start to collect more data, as we speak, it would take decades, if not longer to build up enough data sets to see tangible, meaningful changes in the performance of these models.
Convincing both public and private sector to invest in this is more challenging than what you would think, because the return on investment is not necessarily that immediate, or sometimes that apparent. Say in some context, if you start collecting data, you can sell the data either directly or indirectly and get some feedback or financial compensation. But in this case, it's not going to reward you in a financial way.
But if you're looking at this problem as investing in our strategic resources, I think it is far too easier to justify these costs. At the end of the day, I think the best way I can describe it and our relationship with these models is that AI and computational intelligence should be seen as a tool that we need to supervise, not the other way around.
12:27 - How do extinct marine megafauna measure up?
How do extinct marine megafauna measure up?
Kristina Kocakova, University of Zurich
The modern marine megafauna - broadly meaning big ocean inhabitants - play important ecological roles and include many charismatic species that have drawn the attention of both the scientific community and the public. Just recently, for instance, we learned that migrating whales move thousands of tonnes of nitrogen around the planet. However, the extinct marine megafauna have never been assessed as a whole - in fact there’s not even an agreed definition of what constitutes extinct marine megafauna - and what were their biological and ecological patterns? This was the challenge that, as she explains to Chris Smith, the University of Zurich’s Kristina Kocakova’s set for herself…
Kristina - Large animals in the oceans, which we also call megafauna, are extremely important for these ecosystems. They do a lot of important roles in these oceans, they transport nutrients, they shape the populations of other animals, and so on. And in the modern oceans, we do have a definition for what constitutes marine megafauna, which is heavier than 45 kilograms.
But we actually never properly had a proper definition of marine megafauna of extinct animals, of the ones that we can find in the fossil record. So, the aim of this paper was to do a thorough search of literature and compile a database of all animals that are larger than one metre, and then make a decision whether this kind of value would be good as a definition for what constitutes marine megafauna of extinct animals.
Chris - How far back in time did you go?
Kristina - We went all the way back to the Cambrian, so that's the beginning, that's when the boom of life happened, which is more than 500 million years ago. We actually find our first instances of megafauna already at this time. So, we tried to cover the entire span of complex life on Earth.
Chris - Quite a tall order, that one, isn't it? How did size change over time? Because when we think back to the era of dinosaurs and things, we think of these enormous creatures, where life today is small in comparison. But immediately it strikes me that in the sea we still have today enormous creatures. Blue whales are absolutely huge. And has that always been the case, or has there been a shift over time?
Kristina - When we compiled this data, we've noticed that we always see increase in size over time. Animals do tend to get larger and the rate at which they do it can differ. And it also depends on what kind of group of animals it is, whether we're talking reptiles or invertebrates or mammals.
Some of the largest animals that we found were found in the more recent times, in this period called Neogene, which is from around 23 million years ago. And this is kind of when we started seeing, for example, the iconic megalodon, we started seeing some really large whales. But even though these animals could have reached sizes around 20 metres, they are still 10 metres shorter than the blue whale, which we see around today, which I think is incredible that we happen to live at the same time as the largest marine animal that ever lived.
Chris - The interesting transition that's also, I suppose, happened is that we've gone from an era of largely reptiles to an era where the really big animals are mammals now in the oceans, haven't we?
Kristina - Definitely. So, actually reptiles are the largest group of extinct large animals that we found. The largest amount of large reptiles that we found were identified from the Mesozoic, which is also called the age of reptiles.
And this is exactly the time where you would also be thinking about dinosaurs. We, of course, at the end of the Mesozoic had the bolide impact, which caused all dinosaurs to go extinct. And it also heavily affected animals in the marine realm, which included these large ichthyosaurus, which are these kind of fish-shaped reptiles that really dominated the oceans. And actually the largest animal that we found, not only in the Mesozoic, but overall was an ichthyosaur, which could reach 21 metres in length. In the following era, the radiation of mammals occurred and they started dominating the oceans, although it didn't happen immediately. It did take some time for these large whales and other mammals to spread out and reach the incredible sizes that they reach today.
Chris - Now you've got this, it's obviously a first step to establish a foundation for understanding more about what hasn't made it into the present era. What has emerged in terms of interesting findings and also what gaps has this revealed to us? Things we don't know that we really now need to go after to fill in the holes.
Kristina - This study kind of represents a good starting point just by providing that definition of what constitutes megafauna, what constitutes these impressively large animals, which then can be applied - if needed - in other studies. Another thing is that during the process of gathering information, like for example, the ecological roles. And so, for example, gathering the information of what kind of diet some of these animals had, we realised that for quite a few of them, we still do not have this kind of information. And this is quite crucial to be able to make proper conclusions explaining what was happening in the ecosystems.
We do have some kind of starting point in the study. We do have pretty good information about, for example, the majority of the megafauna were large animals that ate other animals. So, they weren't really filter feeders, unlike what we see today, which is really interesting.
But for many, many other aspects, there are still definitely gaps that need to be fulfilled.
Chris - One thing I often wonder when I sort of go down to the beach and I look out on the sea and realise that three quarters of our planet is covered in ocean. And I often think back to the time when life was in its infancy. Were those oceans absolutely replete and boiling with life? Or were they relatively underpopulated? Was it masses of empty water and the odd thing living in it? What was it like then?
Kristina - It really depends on which specific time in the history of life you look at. Of course, the Cambrian explosion is popular because that was the time where really life was booming. There was a lot of shallow areas, areas where life truly was replete. Whereas if you would go to times after some kind of mass extinctions, of course, life would be more rare.
Chris - So, if you had to go back in time and you were going to take a dip, because I often think, would something eat me the minute I got into one of these primitive oceans? What period would you go back to knowing you'd be safe to have a swim versus if you desperately wanted to get eaten, when would be the best time to go back to?
Kristina - I would say that for a human, I suppose Cambrian would be the best time because that was the era of invertebrates. And I'm not sure if some of these invertebrates would know what to do with a fleshy being such as myself if they found me in the water. Whereas if I would desperately want to get eaten, I would say like anytime when the megalodon was around as a safe bet.
19:29 - Coastal ecosystem effects of climate change
Coastal ecosystem effects of climate change
Scott Doney, University of Virginia
Coastal environments are one of the most important habitats, not just for their carbon sequestration but for their protection against oceanic weather events. The exact impacts that climate change will have on these areas is not yet fully understood. However, what we do have are decades of data, collected by a range of actors from governmental bodies to citizen scientists. So scientists at the University of Virginia have been analysing these data to extract the signals that might relate to climate change, so we can work out what the future might hold. Here’s Scott Doney…
Scott - We're trying to figure out how climate change might be changing coastal marshes and the lagoons that surround them. And we had decades worth of water quality data that had been collected but really hadn't been analysed in depth. So, we were trying to dig in and see: can we see a trend in warming because of climate change? Can we see a trend in the biology of the organisms that live in the lagoon? And then: can we start to understand what the implications of those trends are.
Chris - What specific things were you measuring or subjecting to scrutiny to see if you could see a signal in them?
Scott - It's things like the temperature of the water, the salinity because we're in the coastal ocean and so you have freshwater coming from land and seawater coming in from the ocean. Chlorophyll which is a proxy for how much phytoplankton is there. And then also things like nutrients. Nutrient pollution can drive harmful algal blooms for example.
Chris - How do you tease out from what you see though as a genuine signal of climate change versus something which is a normal cycle which might operate over long time scales but would look like it's climate change because it's changing in line with everything else?
Scott - That is the challenge. The simplest thing that we started with was we wanted to see from this long-term several decades of data could we first just pull out something simple like the seasonal cycle. The study was actually done by a graduate student as part of her master's thesis and one of the things that I thought the student did which was really interesting was there are groups that have also put out autonomous instruments that are measuring very high frequency data like every quarter of an hour but those are only in a few locations so she did statistical tests to see - given that we have some high frequency records - can we sample those high frequency records like the way we're sampling this decadal record and recover the same information. So, it's kind of this idea of we can't go back in time to measure things 40 years ago - but can we say if the record looked like it does now with these robotic instruments can we say something about how the past might have evolved?
Chris - And can you? Does it work?
Scott - It seemed to work. We were able to show at least with the statistics that we had collected enough data over 30 or 40 years to be able to resolve the seasonal cycle and in a lot of places we could actually show that the seasonal cycle is you know normal things warm in the summer and they cool in the winter and what she was able to show in this study is that shape of that seasonal cycle was actually changing so in some places it was getting warmer- but it was also getting warmer earlier in the year and that's what we would expect based on models.
Chris - Is this generalisable? So, could I take your model and take it to the opposite coast of the US or even to a European coastline and apply some of the same techniques in order to ask the same questions there?
Scott - We were hoping to develop a methodology that could be used in other places and one of the really encouraging things - since the paper came out - colleagues at other sites in the US who have this water quality data have applied the same sorts of approaches and are seeing similar results so that's very encouraging.
Chris - And can it capture, because you mentioned we wanted to know what people are doing, the other major dramatic change is in recent decades we've seen the population - the human population - go up by tens of percent. I mean we've got a third more people on earth now than when I was doing my PhD. So does that also get captured by this sort of model and how can we control for that because that must also make quite a considerable impact on a lot of these metrics.
Scott - So, I've worked in other places where there's a big impact from a growing human population one of the biggest signals you see is because of nutrient runoff from fertiliser and from atmospheric pollution and we're trying to use this site in Virginia as sort of a baseline of what signals might look like without that local human factor and then we can then compare that to places that have say a larger input of fertiliser runoff or more atmospheric pollution.
Chris - And when you compare your model with the predictions that other models because other people are doing sort of similar things or they're asking a similar question but in a different way they're approaching it or getting at different outcomes. Does it broadly align with what generally is accepted as the likely trajectory and if so what are the predictions then?
Scott - Our work was consistent in that we expect to see coastal waters warming. The freshwater balance which is really important for the ecosystem is going to vary from place to place. Some places are going to get wetter with more runoff from land and other places are going to get drier so that's going to be pretty site specific.
I think one of the interesting things is we saw at some of the stations that we have data from that when the ocean got warmer there was also more chlorophyll, more phytoplankton and we've seen that in a couple of other places. That's tricky in the models. The models are still kind of looking at that.
There are some places around the globe where we expect the ocean to become more productive and places where we expect productivity to drop off at least where we were looking where we could see what we think is a climate signal we saw warming and increased biological productivity.
25:57 - Drylands: what are they, and what is their productivity?
Drylands: what are they, and what is their productivity?
Lixin Wang, Indiana University
Now from water abundance to the opposite end of the spectrum: drylands. These are arid and semi-arid areas where the amount of water that evaporates or transpires off the land is just about balanced by the amount of water that falls as rain. They actually account for over 40% of the planet’s land surface area, and two and a half billion people live in these regions. But the data we have on them are not comprehensive. And because climate change is almost certain to intensify drought, and increase the planet’s dryland proportion, it’s important that we fill in these gaps. This is what Indiana University’s Lixin Wang is advocating…
Lixin - For this article, we are looking at the complex relationship between plant production and water input in drylands. The reason for that is because drylands have a lot of productivity. When you look at it, it doesn't seem like it has a lot of plants, but actually drylands contribute a lot to the productivity. At the same time, dryland productivity is very important for our food security, for carbon sequestration. So it's very important to understand what's controlling the dryland productivity and to make future predictions.
Chris: And I suppose this is important because we anticipate with climate change that some areas of the world are going to get wetter, but a lot of areas are going to get a lot drier. And so this is almost certainly going to become more important.
Lixin - Exactly. Most of the drylands actually are predicted to become drier in the future decades.
Chris - And is your motivation that there just isn't the evidence base and the knowledge there at the moment? And that's why we need to be putting more attention on this.
Lixin - There are pretty good data-driven evidence showing the rainfall trend for drylands. But for our perspective, for our article, we are looking at the relationship between rainfall and productivity. Because, intuitively, most people think dryland is water limited. If you give them more water, they must have more productivity. This is mostly true, but there are many locations or many scenarios that are not true. So, we basically try to understand why this is the case.
Chris - And why do you think it is?
Lixin - There are actually multiple reasons. So, we talk about the reason from two major perspectives. From the methodology aspect, we talk about what are the potential limitations. We quantify how much water is there. And also we talk about the potential limitations, how we quantify plant productivity in drylands, and what are the potential mistakes or errors we could make. And also we talk about the ecosystem process, the uniqueness of drylands, why sometimes the ecosystem process is responsible for less strong relationship between rainfall and vegetation productivity here in drylands.
Chris - Do they make up a very high fraction of the total then, these particular aspects you're considering? Or are they very much edge cases, and they've been overlooked?
Lixin - There are some factors I think is actually very uniform, maybe across the whole dryland. Some are more site-specific. For example, most often we use precipitation, total annual rainfall, as an indicator of total water input to dryland.
This is okay because rainfall is the most available weather variable we can use. But when you have plants, the water you use is actually from soil. It's not actually rainfall per se.
So there is a translation between rainfall to soil water availability. There are a couple of issues here. It's not all the rainfall can be used by plants, only a portion of this.
Dryland is quite sparse in vegetation, so there are a lot of scenarios where rainfall falling on the ground just gets purely evaporated, so plants can't access that. So that's why some people use the concept of effective precipitation to refer to the part vegetation can use. So ideally, we should use soil water as an indicator of plant water availability. But soil water is not as straightforward as rainfall in terms of measurement.
Chris - Is that what you're advocating for then? Better data, better data collection, a more comprehensive understanding of all of these areas so that we know how to manage them and what to expect from them as we go forward?
Lixin - Exactly. Better data and more coverage of data would be very important. Dryland have been understudied for a very long time.
Most of the time, people assume they are wasteland, there's nothing there. But it's just opposite is true. So, that's why I think the recent effort from different government agencies, different funding agencies, trying to spend more resources, trying to increase the data coverage, but also the type of data we can collect from this very valuable type of ecosystems.
Comments
Add a comment