Reply With QuoteHow did you get the images together into the GIF?Originally Posted by Tiluchi
Your shaded relief looks really good! I'm really jealous that you can get something that looks so good even without Wilbur!
Also thanks for the grayscale image; will be able to get the first run of that going probably tonight
How did you get the images together into the GIF?
For plate IDs, I used the two plates that were rifting apart; this should be covered in the Pasta tutorial. For the temperatures, it was just using Azelor's tutorial, which I've done a few times already so I'm reasonably familiar with. Honestly it just requires some practice, and comparing your results with Earth to see where you might have gone wrong.
I know some people get high tech and use GPlates to export a series of frames, and then stitch those together in movie editing software to create a movie, which they then convert to a gif. I on the other hand am lazy and hate getting too over-complicated with technology, so I just set up a screen recorder, set my GPlates back to the very beginning, and pressed "play", recording my screen. Low-tech, but it got the job done.
NOTE: wrote out a whole post of initial comparisons between my model and the ExoPlaSim and then realized that I was a fool and misread the timesteps. Note for others using ExoPlaSim: timestep 1 doesn't necessarily mean January and timestep 7 doesn't necessarily mean July; for me the hottest month of the southern hemisphere was timestep 4, and the hottest month of the northern hemisphere was timestep 10. Oops!
So, MrBragg was kind enough to run my grayscale map through ExoPlaSim (thanks again! Seriously much appreciated!), and I've been comparing the output with what I got from doing things manually. Since I haven't yet done a manual climate map we won't get to that yet, but the great thing about ExoPlaSim is that it outputs a bunch of climatological maps, including stuff similar to what one would put together with the Azelor tutorial or with earlier versions of the Worldbuilding Pasta tutorial. In case it's useful to others, I'll be going through the ExoPlaSim outputs and comparing them with my own results to see where it differs. ExoPlaSim definitely got some things right where I went wrong, but there are also a number of areas where I'm skeptical of ExoPlaSim's results, most of which seem to boil down to the lack of ocean currents in their climate model.
1. Air pressure:
My initial outputs for January and July are below, along with ExoPlaSim's lower-res (but better-looking) pressure maps.
Rheia_january_pressure.pngRheia_july_pressure.pngpsl_in_rheia01_av10_jan.pngpsl_in_rheia01_av10_jul.png
Overall the results are fairly similar, which is reassuring. However, there are a few differences in interpretation:
1. ExoPlaSim pretty strictly places thermal lows (summertime overland low pressure systems) around 30 degrees, while mine were offset a little based on topography. ExoPlaSim is probably more correct here, and I'll shift them accordingly.
2. While Earth has a semi-permanent high pressure area at very high altitudes (mainly the Himalayas and the Andes), ExoPlaSim doesn't always model this; I'm going to go ahead and keep these high pressure areas as it seems otherwise high mountain plateaus would be modeled as wetter than they really are.
3. ExoPlaSim either understates or omits entirely the overland highs that I had near the poles in the wintertime. Comparing results from ExoPlaSim's earth modeling and real-world data (as seen on Worldbuilding Pasta here), it seems to model polar regions as somewhat warmer than they really are, and I wonder if this is why. At any rate, I'm going to keep my wintertime highs as I want to make sure the areas I imagined as ice caps really are ice caps.
4. ExoPlaSim tends to model subtropical overseas highs as one mostly-contiguous ridge of high pressure, rather than as a series of anticyclones centered near west coasts. While my initial work probably kept high pressure a little bit too concentrated in the anticyclones, from eyeballing the winds it seems like ExoPlaSim kind of understates anticyclonic winds in general. I suspect this is because ExoPlaSim doesn't model currents, and equatorward cold currents are a primary driver of the subtropical highs. At any rate, I'm opting to split the difference and spread my subtropical highs out a little more laterally, but still keep the highest pressure concentrated in the western part.
Overall though I'm pretty happy with how close my initial model was to the ExoPlaSim output.
2. Winds:
Just a quick note on this one, as I don't really like the way ExoPlaSim outputs wind maps. From what I can tell the model seems to strongly favor the prevailing easterlies and westerlies, de-emphasizing cyclonic and anticyclonic winds. I'm not sure if it's a resolution thing or what, but that's in contrary to pretty much any wind map I can find for Earth, where prevailing winds in a given season tend to have a northeast-southwest or northwest-southeast angle, particularly in the tropics. It's also contrary to the example I can see on the Worldbuilding Pasta tutorial, so this is probably user error, though I have no idea how. At any rate, my revised winds are just going to be based on the revised pressure maps, since for the moment I don't trust what I'm getting from ExoPlaSim.
ua_in_rheia01_av10_jan.pngua_in_rheia01_av10_jul.png
With that in mind, here are maps of my revised winds and pressure systems for January and July:
Rheia_pressure_winds_january.pngRheia_pressure_winds_july.png
3. Temperature:
Temperature is where we start to really see some differences between my initial results and the ExoPlaSim model. Maps for January and July are below:
Rheia_january_temperatures.pngRheia_july_temperatures.pngts_in_rheia01_av10_jan.pngts_in_rheia01_av10_jul.png
The main differences I can see are:
1. ExoPlaSim has some really eye-poppingly high temperatures on large continents: it puts average temperatures well over 50 degrees celsius (sometimes closer to 60 degrees C) in the centers of some continents, and has those areas staying above 40º C even in the wintertime. I don't know enough about climate modeling to say why that is, but I'm pretty skeptical of those figures. The model is probably right that I was understating the temperature effect of thermal lows a little bit in my temperature maps, but I'm going to go ahead and not put in the huge swathes of unsurviveably hot temperatures that it's telling me to.
2. On the same token, ExoPlaSim calls for much warmer winter temperatures near the poles than I had; while I have large areas below 25º C in the wintertime (turquoise in my temperature map), the lowest ExoPlaSim calls for is 20º C. Partly that's probably because it somewhat understates the effect of overland high pressure systems near the poles, but also I think the average temperature is just too warm overall. That said, the model mostly agrees about relative temperatures; in other words relatively cold areas in ExoPlaSim's map are also relatively cold areas in my map, so that's good.
3. As expected, the effects of currents aren't really taken into account, so coastal areas with cool currents are somewhat too hot, and temperate areas with warm currents are somewhat too cold. That's because ExoPlaSim doesn't really do currents, and thus I'll keep the effects of currents in my map as they are.
4. Some highland areas are too warm, which is probably just because of limited resolution, as well as the generally very hot average temperature the model is calling for; I'm certainly not going to have 40º C temperatures at over 1,000 meters above sea level. That said, the difference isn't as pronounced as I thought it might be, and again looking at relative levels ExoPlaSim's temperatures and mine mostly align.
With all this in mind, I revised my temperature maps based on the relative temperature output by ExoPlaSim, rather than the absolute values; there are some areas that ExoPlaSim says should be relatively hotter where it's probably correct, but I'm not going to dial it up to the stupidly high temperatures it calls for. I'm also revising them based on my new winds and pressure, though that will have more impact on precipitation than on temperature.
Rheia_temperature_jan_revised.pngRheia_temperatures_jul_revised.png
Next is precipitation, which is probably the most complex part of all this, but I'll save that for another post since I have no idea how long it will take me. Anyway, thoughts or reactions are welcome! This is very much not my area of expertise, so I may well be missing something obvious about climates here...
This is great, I love the detail here! I've not dug into the pressures yet, but your other observations match mine quite closely. Your hybridization of ExoPlaSim and Azelor is going to be a very nice template to follow once I get there
For your winds, the issue there might be that you're plotting one of the uppermost levels of the atmosphere. Panoply by default starts by plotting level 1 (the highest), though for our interests in climate we likely want winds closer to the surface, so maybe not level 10 but something closer to that than 1. You should be able to select the level in the "arrays" tab where you change the timestep.
Alright, now for precipitation, which is probably the part I'm least confident about overall. Unfortunately it also seems to be the area where ExoPlaSim makes the most noticeable systematic errors, so while it's certainly useful comparing the outputs there's definitely lots of room for interpretation here. Below are my first passes at precipitation maps, along with ExoPlaSim's outputs. Note that I didn't take into account orographic lift yet for my July map, as MrBragg sent me the ExoPlaSim results for Rheia before I got to that and I wasn't patient enough to wait.
Rheia_precipitation_jan.pngRheia_precipitation_jul.pngpr_in_rheia01_av10_jan.pngpr_in_rheia01_av10_jul.png
Phew, alright, some noticeable differences there. Specifically:
1. ExoPlaSim moves the ICTZ much less, keeping peak precipitation mostly near the equator regardless of the season. Others have noticed this as well (see: MrBragg's results in his thread, Nikolai's results in the Pasta tutorial I linked in my last post), and it means that equatorial areas are wetter than otherwise expected, and subtropics are much dryer. It seems to be less about pressure systems (which in general I agree with), and more about how it models precipitation and maybe winds. I suspect this has something to do with the lack of currents in the model: onshore winds that pass over warm currents seem to drive the particularly heavy monsoon rains. Still, I'm not sure in general. Whatever the reason, I'm electing to keep my rainiest areas around the ICTZ, rather than right along the equator as ExoPlaSim calls for. On the other hand, I definitely moved my ICTZ a little bit too far in some spots, so I'm moving my heaviest rains equatorward in some spots.
2. ExoPlaSim seems to do a surprisingly good job of modeling lee cyclogenesis (winds passing perpendicular to mountain ranges create some precipitation on the leeward side), and it caught some areas that I missed. I'm including those, and it's an impressive bit of modeling, assuming that's what it's really catching.
3. On the same token it does a pretty good job of modeling the rainshadow effect, modelling leeward areas of mountains as being fairly dry. In some spots I think it might even overstate the effect a little, especially near the ITCZ, but in general it's helped catch some spots I missed.
4.Unsurprisingly, ExoPlaSim doesn't really model the effects of cold currents that make areas east of high-pressure areas particularly dry (think: Atacama and Kalahari Desert), meaning that some areas are wetter than I'd put in. I'm keeping those dry where there's wind blowing off of a cold current from a subtropical high.
5. In addition to the monsoon, ExoPlaSim also seems to underestimate the extratropical storm paths to the west of the subtropical ridge, making east coasts somewhat dryer than they really are; on Earth this shows up as making the southeastern US mediterranean climates and making Japan a little dryer than it should be as well. I probably overstated this effect a little bit in my map so I'm making my east coasts a little dryer, but not as dry as ExoPlaSim calls for.
6. In general it models continental interiors as somewhat drier than what I put; in part this might be due to understating the impact of some of the effects I mentioned above like monsoons and extratropical storms, and this is probably why it modeled Rheia's continental interiors as being blisteringly hot. On the other hand I think I was underestimating the extent to which moist winds would be blocked by all my coastal mountain ranges. Again, I'm making things a little dryer, but not making them into the blast furnaces ExoPlaSim is modeling.
I also just went through and made some other random adjustments where my initial maps differed from the ExoPlaSim model, generally assuming the model was more correct than I was. Here are my revised precipitation maps:
Rheia_precipitation_jan_revised.pngRheia_precipitation_jul_revised.png
I'll leave this here for now, as now I need to figure out how to run the scripts from the Azelor tutorial... did them once already but that was a VERY long time ago. Will make another post when I have my complete climate maps. Anyway, as always comments and corrections are very welcome!
Great job. Though don't stress too much because to have a more precise calculation of climate simulation the tools should also take into account : thermohaline circulation, volcanic activity, solar flares , strenght of solar luminance, albedo reflection , possible meteoric impacts, biological activity density, axial tilt and planet orbit , greenhouse gases densities, in particular co2 concentration, carbon dioxide levels absorbed by oceans relative to its mass, vegetation land cover ... when co2 gradually drops overall it allows the climate to cool, with some ups and downs. Once carbon dioxide concentrations becames low enough (around 300 parts per million) , the ice age cycle begins.
Its all far to complex to take track in precise simulation I think so any good informed guess is more than accurate to me.
A number of these actually are taken into account during the simulation. Solar flux (and star temp), albedo of different land types, vegetation cover, obliquity / eccentricity, and gas concentrations / pressures are all treated. There are certainly simplifications and systematic errors in the model outputs, but in terms of variables that get treated it does quite a thorough job and allows the effects of any one of these variables to be isolated and varied.
Nice then could be an alteration of those parameters that gives abnormal temperatures on continents like I read before?
Posting Permissions
