Best posts made by illili

Re: Wishlist - Model Run Time on Map & Previous Runs

As previously mentioned in the above thread, having the ability to choose the model run (rather than only having a single current run) can be very helpful.

Since Windy doesn't at the moment do ensemble forecasting, being able to flip back and forth between two runs or compare models/layers would be an incredible feature to have.

With the extra real estate on desktop, this could even be a feature where you could run two models/layers side by side to see subtle changes/trends in a storm track run to run.

Easier said than done I know, but would be a great feature to maybe see someday!

@Suty It would be really nice if there was a way to have a "compact" option for the menu. Being able to change font sizes and remove the images for layers to create a text based experience would be incredible.

These menus take up a ton of space on the screen when they swing out, making it really hard to change between layers while also viewing the map (mostly on mobile). Desktop has much more space, but same principle, a lot of screen space is consumed by these large menus compared to a compact text menu.

I know several threads existed where people left the same kind of feedback when this design was rolled out to mobile initially.

Are there any plans to create a more compact layout option? Or is there a more appropriate way to suggest that feedback?

Thank you.

@Suty I see HRDPS in the app now with the most recent update, thanks to the team for adding it!

@Wheats Would love to see this as an option. A “compact” style side menu similar to before and the new option for those who want it.

@Suty Can we please get back the static altitude slider (desktop)?

I wanted to bump this as well. The HRDPS models (1k and 2.5k) are another very helpful tool for (obviously Canada) but big chunks of the US as well. As someone in the Pacific Northwest it's helpful from other areas I can access it. Would love to see it added to Windy as well.

@illili said in Appearance:

@Suty Can we please get back the static altitude slider (desktop)?

@Suty, I'm not sure who's responsible, but now I see there is now a way to use the altitude slider without having to open the entire side menu. I really appreciate that recent addition since the changes.

It's the best of both where it can be independently expanded or compact when not in use. That option is perfect.

Hello,

Currently when comparing models, some regional models "backfill" with other models after their relatively short runs. Some example are the NAM, HRRR and HRDPS "extending" their run with the GFS and ECMWF respectively.

It's marked so it's fairly easy to tell, but I don't see the value in the feature when the compare view has the longer range models in it regardless.

It would be helpful to be able to optionally disable it to clean up the view if chosen.

@tessellahedron For a general purpose app, I like NOAA Weather and Tides (the icon is a sun and cloud in a crosshair and is free). That more or less takes what's on NOAAs website and puts it into a simple app. It lets you pick a specific spot on the map and will give you the forecast for whatever grid it's in.

But for Windy, yes I like the visual map most since that shows any given layer (wind, gusts, rain, temp, etc.) and lets me see what's most likely to happen based on the surrounding conditions of my area.

In Windy, you can get the forecast for a given location to get a more generalized readout. You can fiddle with the options to pick a different model, or if you want to compare you can set it to do that as well. Then it'll show each model output on a single timeline so you can get an idea if there's consensus or not. The Meteogram option has what you want with forecasted average and gust speeds.

@Suty The issue resolved itself sometime that evening I think. When I checked the next day for the 12z updates it was working as normal.

Sorry for the delay!

I like the new “crosshairs” and associated output when scrolling around the map.

As the OP said, I’m not a fan of the tiles for the sidebar menu. I added favorites but I thumb through a lot of layers so it’s quite slow to find what I need since when the images are so small they start to look the same.

Another poster said it best in my opinion; leave both options available. A “compact” option, if you will, that’s text based like the older style and then also have the tile system for less heavy users or those who simply prefer it.

Also, the new sidebar menu takes up so much screen space. I can no longer slide through from surface up through the atmosphere to get an idea of wind. I have to stop at each level, then close the menu to see. Then open it back up again and repeat.

For all my comments, I still love the app and appreciate all the hard work you all do!

Last update was a while ago:

I just wanted to say thank you for adding this model! It's very helpful to round out the forecast. Now, I have a tendency to look a gift horse in the mouth, is there any chance the HRDPS 1km will end up being added? I know from previous posts it's a big lift. Just curious.

Thanks again, love this addition!

Edit: I didn't see it on the App yet, I assume that's coming soon?

Just wanted to say thank you to the team for adding this to quickly differentiate!

A very small change, but makes a big difference for me as I'm constantly using the North America outputs and not the AK as much.

Hello,

A few updates back, the model selector on desktop took on a compact view. Don't get me wrong, I love compact views. However, as implemented it adds some extra clicks and I think could be improved with some tweaks.

It would be nice if we could rearrange which models show in this compact view or alternatively "force" the model selector to show all available models for the given viewing area without automatically collapsing.

Thank you!

@tessellahedron 7-10 days is not incredibly accurate with any model or system in my experience. At that range, you typically only have lower resolution models available that can provide a general idea, but rarely will be able to predict with precision what is going to happen that far out.

To get a very localized forecast that's as close to accurate as possible, you have to synthesize lots of data sources. For example, combining medium range forecast models (ECMWF, GFS, ICON) with the higher resolution models as things get close (things like HRRR or NAM). You can also sometimes find other higher resolution models that might cover your area to help in this process. Throwing in pre-made forecasts from major sources like weather.com or accuweather can also be helpful as a grounding point.

There's also the "reading between the lines" aspect to consider, where over time you'll see that certain models and weather patterns have a tendency to over/underestimate conditions for your hyper specific location when represented as a forecast.

@jamesflint I use Desktop as my preferred method. Surface pressure isobars are just above the model picker:

Geopotential height unfortunately is buried a bit by swinging out the more layers menu:. Oddly enough you can switch between isoline settings more easily on the mobile app.

Re: Revisiting ensemble options

I see an old thread that was referencing ensembles, is this a feature that is still on the roadmap for development?

Thank you.

@idefix37 I cleared local cache and restarted the browser a couple times. Didn't work the first time, but is now.

Thank you!

@tessellahedron For the maps, it's a matter of preference but they are also useful as you get to know your area better.

For example, if I just look at a graph output that says it's going to be a south wind of 10mph, that's helpful but if I look at a map version of a forecast, I can see much more clearly how likely that is to happen given the current setup.

The reason is I can see how a particular storm is "shaped", what is the tilt of the pressure gradient, what are the winds above the surface doing (those can enhance ground level winds depending how they setup, or steer a storm away), will a rain shadow likely form, etc..

All this amounts to me being able to make a better guess at a hyper local forecast of wind and rain.

Ultimately, no single source is going to be right all the time. For the most accurate forecast, you need to combine multiple models and sources of information. It's why professionals also use ensemble forecasting, where they run a model multiple times with slight variations to average out the inherent shortcomings of a single model run.