 Okay, yes, thank you everyone who has joined us so far. My name is Simon Clarke, I'm EGU's Programmes Coordinator, and welcome to this webinar, How to Visualise Your Research Using Scientific Accessible Graphics. You'll be learning how to visualise results in a manner that is true to data, how to emphasise the readability of your results, including how to make them accessible. The webinar itself will last approximately four to five minutes with a Q&A session at the end. If you have a question, please type it in in a Q&A box at the bottom of your screen, where you can also upload questions you think should be answered. So to take us on our journey today is Fabio Cromeri, who is a freelance researcher and graphic designer of science related content, and this includes producing scientific colour maps for data true and inclusive visualisation. So welcome Fabio, would you like to begin? Thank you Simon, yes. Okay, hello everyone. I hope you're already, so this is going to be a lot. Unfortunately, one webinar is not enough to tell you everything about graphic design. As you know, graphic design is a scientific discipline on its own like geosciences. But since we as geoscientists are no experts in graphic design, I thought we'll be good to just cover the basics of graphic design. So what I will talk about is try to explain how to create effective graphics that excel your research. How to make them scientifically accurate so that they don't distort the underlying data and how to make them accessible to all your peers and the viewers in general. Most of the information you will also find on my web page, fabiocromeri.ch, and I think we're ready to just dive in. And I would like, of course, to start this webinar with graphic. You've all seen this and you all know what it means, even though it's simple, it can communicate a lot. So what's this one here, and quite the opposite manner, and this one as well, this one as well. So visualization is really powerful. If you would show this one, or this one, this wouldn't work in this quite the same manner. And so this is because it has been done right it was either too complex or too simple. So visualization really must be done right. It must also be accessible to everyone. So if you couldn't read the text to the left and the right because it was too small. For example, then visualization has failed and your figure might have failed as well. Visualization and graphics like the heart are super powerful. Since they can span generations and cultures. You only need to think of the emojis which really has become a global language now. It's all graphics basically. Also for science, this is quite important because without nice figures. You won't see good science either. And this is especially true today where research has become an everyday competition for everyone's attention. So so many paper are published these days that we often don't have time to read all of the papers we would like to read. So what we do is just read the title and look at the figures mostly and if the figures aren't conveying your message, then you often lose your readership already. So know the importance of graphic design know the basics and also know when you need professional help. As I said visualization is probably one of the most widely used scientific methodologies we use in research, which is great but at the same time it's also probably the least widely taught one, which is a huge shortcoming so I'm throughout my whole education as a geoscientist I never had one single class teaching me how to make figures and or just the basics about graphic design. So we really are not experts here. So this is why I give the seminar. And this is a short overview or what I will talk about. So first I will start to explain you the different flavors of figures. There's a difference between individual figures, then show you key visualization pitfalls that we have to avoid. And then show you over all the visuals visualization tools images typefaces and so on, and then end up by giving you or trying to give you a simple recipe on how to make a good figure. So, there are different flavors of figures. And for scientific figures, these are mainly artistic impressions conceptual illustrations and data visualizations and they have all different aspects that we need to get right to in order to make them work. So artistic impressions basically represent a fixtures view on a scientific concept like here, as shown on the right hand side, this is a simple representation of the earth, not everything is perfectly represented but it gives you an idea of how something might look like. If it were the case that we would extract some pieces of the earth in this example. Then we have conceptual illustrations like this example here about ocean plate tectonics. And these are mainly freehand drawings, not always freehand but drawings that can portray a certain concept qualitatively. So here we highlight the oceanic plate as it is dragged down by it's in blade portions by the slab pool. And it gets formed in some places on the surface then cools and then gets destructed in another place. So this can kind of convey this concept of ocean plate tectonics very nicely. And finally there is data visualizations. Also very often used graphic. These are basically database maps or charts or graphs that really quantify now the data. And they do it by relating the online data via graphical rulers or scales. So here, this is really key that the mapping of the data is correctly done. Since there's not really one single way of making a good graphic. It's important that we just don't make any major mistakes so preventing pitfalls is kind of key. And I guess most of you know some visualization pitfalls. So if you think about what could be some pitfalls you might have some in your head. These are some I could come up with, for example, perceptionally non uniform color maps that distort the underlying data, and which are actually also faulty scales. And then for the scales in general inaccessible color coding, coding that is not readable by some people because of, for example, red and green combinations. No scales at all is a common one as well, even though we as geoscientists have been hammered in that we should use a hammer or a coin when you take pictures close to rocks, just to get the scales right. And then in general inaccessible graphs like 3d pie plots, for example, which I'll come back to unreadable annotations, if the text is too small and readable, if the figure in general is overloaded. If people try to get in all kinds of information and end up basically communicating nothing at all. Or if the graphic designer fails to really reach the audience. And then finally, also, what often happens is that people start miss to give the acknowledgement to things they reuse in their figures. All these can be grouped into kind of critical pitfalls and also pitfalls that are not critical per se but should be avoided. But the critical ones on top they will really have to avoid them. And, even though that is currently widely not the case. Just to highlight a few of those. So misleading graph types would be a 3d pie chart can be really misleading. Here you see Steve Jobs during an apple keynote, and he actually knows about the flaws of this graphic design, and he uses it to his advantage so he shows the apple share on the green green pie piece that is closest to you, which looks much larger due to the visual distortion then for example the purple one in the back, even though factually it's smaller. And this is because pie charts are difficult for us because we cannot compare angles easily. And also the 3d view of course makes things look bigger, closer they are to you. So we can't use that in science. Another example is bar plots without the zero baseline. This is an example where you, your intuition is that this data set increased a lot during the past years, which is actually not the case if you look at it in a proper graphic. Then you see that the entire data set basically didn't change at all. So you should always put a zero baseline for graph plots. This is basically used by politically inclined parties. Like this one here where basically they don't use a zero baseline to just try and highlight that the prehension order apprehension increased a lot during the last years, which was probably not the case. So they go even further and actually kind of truly mislead the viewers here by just distorting scales. So at the bottom you have a difference of 30 on top you have a difference of 50. And this makes, you know, all these cases, I think it was COVID cases, and look as if they are just gradually increasing instead of a much stronger increase. They go even even that far to make some data points look much lower than they actually are by even further the squeezing the axis. Of course in science, we can't do that as well. However, we often do that as well in science. So you all know rainbow color map. So here on the left hand side you see a scientific version of the data. Basically a gradual increase towards the center with some fluctuations. If you represent it with a full scales like the one on the right hand side, you see all kinds of visual artifacts and then take tired data set is misrepresented. So this shouldn't be used as well. Finally some inaccessible graphics. If you have some line plots and color them with like red green and blue, like on the right hand side, and this is hardly visible for many people. And also when you turn it into a grayscale graph or for people color blindness, then this figure becomes unreadable. The bigger but you need to have is the connection between the the graph individual graphs and the data set marked here as ABC. And if you have it separated like this and just connected via the color coding, then this can fail. So to create science graphics we need to know our visualization tools. I'll show you a few of them and I will now go through all of them. So these are images type phases graphs, graph scale and access reference frames and then finally loads of colors. And then one more thing at the end. So first up images. Try to use high resolution ones. If you use images that they that are pre existing really check the licensing and site sources. There is a lot of open source. Open access images on the web so useful sites are on splash.com pixel beta com or for us very relevant and useful is image.io dot each you dot you, where you also can upload your own images. And then another tool is type phases, which probably most figures use as well. I remember is that type phases have voice so they they also communicate something. So it's good to to think about these as well. And then of course it matters whether you use a type phase for a title for a paragraph for digital or print, and also whether you want to choose clarity or readability. So you would use some series funds like alphaitica, arial, futura or open sons, rather than series funds that you have seen at the bottom. And again, you have some pre installed font type phases on your, on your computer, but of course you can go to the internet and download more open funds, for example from Google the font or font square. There's an example figure where I used open sons. I really like this one as well, because it's kind of clear. And this is just a subduction zone initiation event. And we represent the rocks that are placed on top during this event. And then I used different funds, bold and italic to make to make some of these words pop out more. So I think it's a very clear font that can be used online as well. Another tool is of course the graphs that we represent our data in here I just show you an example of all the metal plots that are possible to create. In general, I would suggest to just go for the easier version, rather than the more complicated ones, and try to avoid the problems for example the 3d pie charts. If you're interested in more complicated and elaborate examples. Say not graphics is a nice webpage where you can see them, but this is really done by graphic design experts so you really have to put in your effort to get these right as well. So, keep it simple if you can. I really like our plots. I think they are simple and effective. Because they allow you to really quantify the differences between individual data sets so you can really tell whether it's one bar is half of the other or not, whether it's a bit less or a bit more. And if you do it like this you can also show the total amount of events in this case, versus the actual counts with the darker pink color indicated here. They are easy to look at and represent the data properly. Graph scales, they are used to make relative proportions accessible. For example the size of an object or the duration of a period. So it's key to include them wherever you can. These are some examples where you don't need an actual scale on it. For example if you show the earth. I guess all of us know how why did this. So this is kind of an implicit scale already so these cases we don't need the scale. But in most other cases, the size or the duration is really important to know. So an important scale is also graph axis. And they are used to map data to a position. If you have a position axis like X and Y, or to a color if you have a color bar, and then also done dimension they need to be correctly represented so people often use abbreviations instead of dimensions for example for the, for the variable year. People use Y RS, which is not a proper dimension and can cause confusion. I actually wrote a whole blog post about the theory interest. Graph axis are really important as this example shows so that's a geologic timescale. And often when we look at geologic timescale. We have squeezed axis, just because we know much more about the more recent earth, like since 800 million years or so. And then we know about the earth for the back in time. And, but representing it sometimes with a proper axis can really help us to, to, to appreciate how much time has passed before them. And then of course we come up with names like the boring billion and so on. The reference frame is another important tool for each graphic. It's used to provide a perspective on the data so this can be 2D versus 3D. It can be different map projections, or it can be the canvas where you put your figure on. For example of map projections, you fall into all the different kinds. And sometimes, if you don't know what they stand for it, it can be a bit too much to, to decide on one. So there's basically three different flavors of map projections. There's an equal area map projection, which represents the area according to nature, and more why there is an example of that. So if you want to show the distribution of something on the Earth's surface, you would probably choose more wider for that to not misrepresent your data. Then there's something called conformal projection. Where the local shape is preserved, as for example in the Mercator projection. So if you really want to see how Iceland looks like geometrically accurately. Then you would probably use this map to do that. And finally, there's equidistant projections that preserve the distance across the map. So if you plan to travel from the north to the south pole, you probably would use a map like Cassini to really know how far you've come once you've reached Africa or something. So it's good to think about the map projection each time you represent a new data set. And then the canvas can be important. So in the middle you see an original image on white canvas. Then if we often represent it on a dark background, so here it's just black, and you can do that by just maintaining the colors on the left hand side. This however reduces the effectiveness of the figure, because now you have the strongest contrast. It's not the darker colors, but the light colors. It's because of the background you've chosen. And this is actually the most boring parts in this figure. What you want to do is like on the right hand side, highlight the features to colors. And you can do that by invert all colors. What you have to be careful with is just not to simply invert the colors because then you also change the hue, but make dark red to light red and light blue to dark blue and so. And then color, which is probably my favorite topic. It is often used to make the figure content more accessible. But it's really important to keep in mind that you should keep it to a minimum. And put it, if you put it differently, and as some other people say this, gray is your best friend in data visualization. So try to start with the figure in gray scale and then add color where it's really necessary. There's just an example of a figure. This is actually a figure showing how to be perceived color. And here I found it necessary to color the light, the different wavelength of the light from the light source to the colored object in the respective colors because you have the blue wavelength, the green wavelength, the red wavelength. But now you have to be careful because some people cannot distinguish some of these color, for example green and red. So you have to make sure that your figure is also readable without the actual color coding. And I did this by just having differently spaced dashes for these lines, which actually also represent the short wavelengths and the long wavelength. So the figure content is still perfectly readable for all viewers. Now you can go even further and use more colors. Here's again a geologic time scale, but this one is actually color vision deficiency friendly. So I use all the different discrete battle colors to color all the different times in here. Now you can do simulations of how it might look like to some people. So this is Deuteranopia simulation for green blind people. This is Protanopia simulation for red blind people. This is Tritanopia for blue blindness. And then even for acromatopsia, which is nothing else than grayscale for color blind people or if you print it in black and white. This is still readable as is. Of course, the more color you use, the closer they become to each other and get harder to distinguish. But finally, you don't have overlapping colors at all. So all these color palettes, of course, they are really difficult to create. So I provide them on my web page. So instead of the scientific color maps package, I also recommend read some very excellent online blog posts by Lisa Charlotte Muth on colors for data visualization and she is an actual expert on this. And then if you want to know more on color, I give entire lectures like this one here just about color. Another important topic related color is color gradients. And we use them a lot in data visualization just when we do color mapping. So we highlight this now a little bit. But first we need to understand what color mapping actually is. So if you have a 3D graph with representing three dimensions x, y and set. Make this much simpler to present by just using a color bar. So if you turn the third dimension from represented in position to a representation in color, we can actually squeeze the entire plot to a 2D surface, which is useful because we often present these data sets on a flat canvas like a screen or in a paper. But now we have to be careful with the properties. So the color here represents the same as the space does in the axis on the left hand side. And no surprise that there are really bad color maps are factually bad. It's not just my opinion. I'm like the rainbow, which is one of the most used collapse during for in research general. It fails because it represents a change in data differently along the color bar. So for example a change between one and two should look the same as a change between around six and seven but it doesn't. There's a huge difference and misrepresentation and it's not surprising that most people see, you know their features at the boundary between yellow and red. To make that even more clear you can actually represent it as a position axis. And this is the exact representation of it, you have, you know, squeeze the space instead of the color here. And I think it becomes clear to every scientist that you wouldn't be able to even think about publishing a graph and axis like this, but for some reason the color map that is faulted is still often accepted for publications. Luckily, some clever people found out how to create perceptionally uniform gradients. This is what we need. We need even color gradients all along the entire gradient. So people have developed this perceptionally uniform color space. And one is called CCAM02 UCS. It's very complicated and a lot of research has come into this. But what you need to know is that we can extract a color difference metric called Delta E, which basically gives the personal difference between two individual colors. So how we perceive the change in two individual colors. So this is very useful because now we can actually quantify the change in color along a color gradient, and we can also develop actual good ones. So on the left hand side you see Betelow, where the graph is flat as it should be. So the change between neighboring colors doesn't vary along the color bar, whereas for chat for the rainbow color map it varies a lot. So this is a bad one. And now you can actually even quantify that's not my opinion again. The error this introduces to your data set, and this is negligible for Betelow, but can be more than 8% of the data range shown with the color map to your data. So this is likely the biggest error in your data set. And it can easily be corrected by just using a proper color map. To further highlight what you do to your data if you use a bad color map, we can now quantify also the representation of a flat slope. So if you represent a flat slope with Betelow it remains flat. If you represent a flat slope with chat it starts to be wobbly and you wouldn't have an idea that it initially that it actually is flat. So, luckily, there's some good factually good color maps. For example, the scientific color maps version 7 that you see here. They're all perceptionally uniform and don't distort your data visually. They are perceptionally ordered very intuitive to read their color vision deficiency friendly. And don't exclude your some of your readership. They're even readable in black and white. What's more, all of these color maps they they have different types that we need to use for different data sets and the scientific color maps, all of these different types and classes are provided. So you also have continuous ones you have discrete ones and you have even categorical ones. They are compatible with all major software packages. They're now provided in more than 20 different formats. They are versioned and citable, which acknowledge the work that went into them, and they are free to use. So they're really the scientific choice here. And you can find them on my webpage. So let's have a quick breather, because there was already quite a lot of information in quite a short time. But so we've seen a lot of visualization tools ranging from images typefaces graphs, graph scales graph axis, the reference frame and color. But there's one more thing one more tool I haven't mentioned and it's quite an important one. And it's actually here. And it's also here. It's empty space and empty spaces is super critical and useful. It's used to make the figure more clear. And in some cases even convey information. So it's really extremely powerful. Really. Can you see how important this really looks like. It's just because of the empty space. This is an actual figure where I used empty space to convey some information. For example, politicians are profiled elaborate profile athletes have an elaborate profile to understand them. Academics have one single number that ranks them. This number is even flawed and biased and unfair. So I show this by having a lot of empty space around it and this single number in the middle. So this empty space really highlights the number, even though it's small, it kind of pops out more. So these were all the tools you need to know. Now we want to make a good figure. I tried to give you a recipe here. You can start by knowing what your message is know what your audience is and knowing what the figures environment is. Then you have created the first draft of a figure for example this one here, then good way to to continues to perform the highway billboard test. Now you can imagine your graphic on a highway billboard and the people driving by looking at it have around 10 seconds to to get your main message. And not more. And this already helps you to design your figure in a better way. So this would be a better example here. But now how to make this better example and there is really the graphic design principles come in. So it's good to know all these different design principles. For example, there is contrast, there is symmetry or asymmetry. There is graphic hierarchy, continuity and accessibility. And now we quickly go through all of them and have a look at them is some examples as well. So contrast this is a figure that uses contrast in quite a nice way in different ways. So, just by having a light gray color in the back you basically give the context of it it's the earth, a map of the earth surface. And then you have this high contrast blue blobs that contrast not only the more complex shape in the background, some simple rounded shape. They also contrast the entire figure with a color that helps the most important part to pop out most. Then symmetry is used to make the figure much more friendly to look at often. You can also use asymmetry in some cases so this is an example showing all the major elements of the earth mantle. The areas represent the actual amount of these elements, but they are also arranged in a way that makes it pleasant to look at. So it finally makes your figure also more effective. Then we have hierarchy. This is a very important one. So this is a super complicated figure and often you would say this is already too much, but if you put in some hierarchy and guide the viewer's eye, then you can make it work. So first, you will probably look at geodynamic modeling so you know what it's all about. And then your eye would be guided to the to the arrow that goes from nature to scientific progress and the white bars on top of it. Then once you got all this, you go to the gray text to get more in depth information. So if you do this, you can make even complicated figures work kind of effectively. Then continuity is also a nice one. Here is a figure with three different panels, but we have a continuity throughout the entire field. So we have, first of all, the same typeface across this figure. And then we have the same coloring across this figure. And we have the same shapes across this figure. So just by continuity, you know that a horizontal bar, for example, on the left represents a noceanic plate. This is just because you have similar shapes throughout this figure. So we have a gradient in color, which represents the timing. So we go from light gray to dark gray. And this is also applicable to all the three different panels. And the same goes with the fold, which is colored in pinkish colors. That also gives you, on the first site, an idea where how the system moves and where the important beats are. And finally, accessibility. Of course, you need to make sure people can see and understand and access the information. This is a periodic table of elements, but also now in scientific coloring that is color vision deficiency friendly. So you can actually know which elements are in gas form and which are in liquid form, for example. And some people might find it hard to read this typeface. So we on sing.org, where we provide this figure, we also provide a figure that has a more friendly typeface, which is healthy in this case. So it's also important to think about, yeah, is the figure accessible or not. Now I've showed you how to make a good figure. Now, of course, we all want to make a great figure as well. And there, the only tip I can give you is really to put in the work and iterate obsessively. So if you have a final figure and do another round of improvements, another round, yet another round, then submit your figure and maybe resubmit it again. Because only then you will really have a figure that is really great and you like. Once you have one of these great figures. Please consider sharing it as well. For example, on the new graphics platform. It's called sync.org. It's from source to ink. And it's it's a geoscience related. So it has a lot of figures that are commonly used throughout our discipline. And you can upload your own figures. You can clarify the licensing of the graphics. And make them citable by uploading into some report repositories like Zenodo and to make them more widely useful by providing not just one figure but alternations of it as well. And finally, of course, they can be easily findable than if they're all in the same place. And this is just a quick overview over the sync.org platform. It has a powerful search engine. Some keywords you can find some common graphics there. Then you can also contribute of course your own graphics that are reviewed by an expert team, but also by the entire scientific community, you can comment on all the figures. And some some important guidelines and how to create graphics. And, you know, finally a guide like a gallery where you can look for figures if you need some. So, I just have time to kind of summarize and I will do this with going through making a great poster or should have been a display this year because you just general assembly will have displays this year. But we'll focus on the poster. So if you make a poster design, first of all, we need to understand what's a poster and the poster is a quick and concise view of your research. I'm quick and concise so it's not meant to be copy pasted text from your paper onto a sheet of paper. Again, remember, we designed for a purpose. We need to know our message. So for example, we found this and that. And really, it helps me to really write those three things down before I start to create figures, because otherwise I just don't think about it and then my figure isn't effective enough. If you didn't know your audience this for the general assembly, this would be a broad range of geoscientists, which means experts and non experts of your particular field. And know the figures environment which would be a poster board in old days, maybe also future days in a big poster hall. In this, there's actually a super nice design, which is called the better poster. And it's designed by Mike Morrison. And on sync.org you'll find a template for PowerPoint and keynote. And it looks like this. This is a clear font that is easy to read. It has simple elements so it has this main plate to make a main message in the middle, which is separated on the sides. It has small text boxes or figure boxes. It has an empty space. And again, there's a division into macro empty space and micro empty space. So micro empty space is really to give emphasis to something and micro empty space is used to, to group different elements. And then it uses contrast. The middle panel really pops out the main message from the more detailed parts on the sides. And it has symmetry as well. It has hierarchy. Your, your eyes guided to the important bits. So when you walk past this poster you, you know exactly what it's about what the main finding is. So if you're interested, you go in and read the title and, and all the details of the results. So it's an effective way to have a poster. And the best of it is that this kind of posters created in a very short amount of time so people always say they don't have enough time to make a nice poster but it's not the case with this. And this is clearly one of the most effective ones you can make. So yeah, use it. So to conclude. Remember that there are different flavors of figures and they all need different aspects to get right. Avoid visualization pitfalls, especially use proper scales use colored bars with scientific color gradients. So your figures for accessibility, remind your peers if they don't and teach your students that it's, it's key to do that because we still have, we still don't teach to assign students about graphic design. Then know your visualization tools with gone through a couple of them, follow the billboard test and apply some graphic design principles. And then you will get towards a great figure, which you then can use our share on an online platform with community, which will be great on sync.org. So, thank you very much for listening until the very end. Just remember that visualization deserves all your full attention, and good luck creating the figures for the general assembly and your research in general. Thanks. Excellent. Thank you, Fabio. One of the key takeaways there is not to use a rainbow color map. Yeah, so we have some time for some questions. Before I jump into that, I just want to quickly say that EG webinars are published a week after on our YouTube channel. So type into edu into YouTube or the link is also in the chat. And this time next week, we have to be watching and use it as a resource. So just moving on to some of the questions. We've got quite a few, but we'll try and get through as many as possible. One is from Frank Swan, and it's, I think it's two questions in one. One, and I'm just rephrasing here. And it's basically, is it possible that graphic design can be too flashy. In fact, it might be off putting. Is there a way to show some posterity, for example, if you think back to profit design in the 70s, isn't really seen as so attractive in the current era. I don't know if I had any answers to that as well. So the first part is really important because it really guides our eyes if there's something that attracts the I strongly like bright colors. Then it becomes really a way to guide our eyes. And this can be a problematic for data visualization if with one part of the data is attracting the I more than the other. Basically warm color do that more than cold color. So it can be a bit of a problem if you use both warming cold colors in one figure. Regarding the other part of the question. I guess this then becomes more discussion of decoration and what we personally feel about certain color combinations. So it becomes less a question of science or scientific graphic design. But yeah, I mean, I don't know, just try to make all the nice figures I guess it's very useful as well. So to predict how trends might change, but I guess the first step is always to make sure that whatever data you want to represent. It's using, for example, scientific maps rather than one space necessary on current aesthetic trends. There are a couple of questions asking about asking for program recommendations. Do you have any for graphic design, in particular any tools for 3D or 2D conceptual figures. There are a couple, of course, first we always highlight the open, openly accessible ones. These ones are our GIMP as a terrible name, but it's useful. And then there is Inkscape, which is basically the open version of something like Adobe Illustrator. Then there are others. I actually, until recently, I often used Microsoft PowerPoint or Keynote to create graphics because it's super efficient to do that. And if you just want to put together some shapes and text, of course, you cannot export them into a vector design graphics then. But for many cases, this is enough. And then if you go towards data visualization, of course, then you go into all, you can code everything on your own using Python or MATLAB and so on. You can use 3D graphic design programs like Visit or Paraview. If you have big three-dimensional data sets, they are super nice to use. And then if you just do graphic design by hand, I really like the Affinity design software, which you have to pay a bit, but it's quite cheap. It's cheaper than Adobe Illustrate. And apart from that, maybe also pen and paper, you can always digitize it afterwards. Thank you. Another question is asking, it's not possible to know what level of color blindness someone might have. So it's asking, how do we go about choosing the color maps? And is there one that can best use to accommodate the most people or not? Yeah, so all the scientific color maps that I showed you, they are readable by all people because they are readable in grayskate. So the simplest test to do is just look whether they are readable in grayskate. As I've done, all the scientific color maps, they have a lightness gradient, which makes them readable to all people. And if you're unsure about the color map that you find in your favorite software program, you can either convert it to grayscale or on the web. Like there is some tools where you can check for color blind simulations. Just Google color blind simulation and you'll find a web page like Kobliss that provides you with the simulations of your figure. That's just uploading the figure. So that's a simple test to make sure that it is readable is check it out, check if it can be read in a grayscale. Yeah. Any questions asking for recommendations on courses, but if they want to dive a bit deeper, what resources can they go to to explore a bit further? Well, for the easiest way to do that, or the easiest. There's a lot of blog posts online. You don't always know whether the resources is done by an expert, but you can get a lot of information but just following some graphic design experts and see I've published some blog posts, so there's a lot of this online. And other than that, yeah, contact graphic designers. I also give presentations on, for example, on color specifically. Yeah, just go to the experts, I guess. The experts find their content there, putting out there. So, earlier in the presentation, you did recommend a few places to go. So, next Wednesday when recording is up and you want to look further. Feel free to browse this presentation at a more leisurely pace or resources. What was sent in to me was asking about. I do have any thoughts on using colors, which champion sending a message over perhaps that's scientific and kind of to work together. I'm thinking, in particular about the NASA figures NASA, representation of ozone depletion in 1980s where areas of ozone depletion where there's fine angry red, which conveyed as per its mother obviously around the situation. Is it, is it possible about those two can work together the need for to convey a message and keeping color design scientific. Definitely, I think that's one of the reasons why I provide so many different scientific color maps that you can really choose colors that are most optimally representing your message, and not just the data. So, as long as you use a scientific color map that truly represents the data you can choose the colors that highlight some parts more. For example, if you have a white background, and you have a data set goes from zero to 100. And you want to show the values in 100 are the most important ones for example, danger level 100%. Then you would use a color map going from some light color, because it has a low contrast with the background to, you know, some reddish color that really makes you think there is something there. So I can definitely combine those two. Again, it goes back to those core principles of contrast and hierarchy for thinking how to push forward and message was keeping it some today. Yeah. Another question is to have suggestions for organizing multi panel figures. So I'm guessing that the focus isn't just on the single image, but multiple. Yeah, I guess I'm applied the same design principles make sure there's continuity represent, for example, the same data set with the same colors that even applies throughout your entire paper I would say, if you have a publication show CO2 throughout the paper then always use the same colors for this variable. Yeah, continuity use the same font, use the same symbols, and just make sure you don't overload the entire figure because when you have multiple panels you can finally be quite small and put on a page on a A4 page. Yeah, and otherwise just put them as single figures. Sure. Thank you. So we have time for one more question as time is quickly running out. And that is, do you have a favorite published figure at all, or anything that perhaps you can recommend and people can look at for inspiration. Um, but yeah, I think most figures on the sync.org platform are really nice so you should go look at them. But if I had to pick one single figure I would probably go with something like the warming stripes. I don't know if you know them but they basically represent the change in global mean temperature over the last couple of years showing global warming quite dramatically and in a very simple way. So, simple figures that have an important message are my favorite I guess. Excellent. Um, and that marks the end of the webinar. So let's say thank you Fabio for a really informative and useful presentation. It definitely got me reflecting on some of my past attempts at visualization. Thank you for everyone who joins and the questions before. As I said before, all EG webinars are uploaded to our YouTube channel a week later. So if you want to come back to this as a resource, keep an eye out. Otherwise, thanks a lot, and I'll see you in our next webinar. Bye bye.