How to Use Windy.com for Landscape Photography

Versione in Italiano QUI

Introduction

I would undoubtedly rank meteorology at the top of the things that make landscape photography both challenging and captivating.

As an uncontrollable entity by definition, the weather is the true god of every photography outing: no matter how thorough our planning, it can determine the outcome of a shot up until the last moment with a snap of its fingers and a mocking smile (that’s how I imagine every photographic failure).

How often has the perfect evening turned into a failure because of a cloud suddenly appearing on the horizon, or a rainy, stormy day ended up gifting us with epic light at the last second?

The only constant in these two situations (and all those in between) is that we cannot control the weather like a studio light. Period.

This leads us to one of the two possible categories that divide the world of landscape photographers: those who suffer the uncontrollability of the weather during every outing and passively accept it as fate, and those who believe that, although the weather can’t be controlled, one can try to learn its language to understand when (and where) it’s worth taking a chance and when it’s better to stay home and open a good bottle of wine. (It would be worth mentioning the “oh, you’re always so lucky” crowd, but let’s just leave that aside..)

This (long… perhaps way too long… sorry) article is aimed at both categories: those who want to try to determine their own destiny (at least in photography) and those who are already trying but with the same success I had with Latin in high school (and the fact that I ended up studying Engineering should tell you something).

Disclaimer

This article was not written in collaboration with Windy.com in any way, shape, or form. Even my Premium subscription is paid out of my own pocket. 🙂

The article ended up being infinitely longer than expected, even though I realize I’ve shortened it quite a bit.

This subject is already complex on its own, and when applied to photography, it becomes even more complicated because, as far as I know, no one has ever thoroughly addressed these topics. What you’ll find below is, therefore, solely and exclusively my experience, offering you some practical tools for your upcoming photography outings.

Finally, as always, this article doesn’t contain absolute truths but merely my personal thoughts.

Why Windy.com?

There are many ways to fail when it comes to weather.

The simplest and most obvious way is by reading the wrong forecast: with so many apps available, we often get swayed by the one with the friendliest graphics or the most advertised at the moment (not to mention that about 70% of these apps source their data from the same place and interpret it in their own way).

Another way to fail, which I personally find amusing, is to check multiple weather sources and then choose the one that gives us the results we prefer or believe the one that the majority predicts.

After trying all of this, I realized it might be time to study the matter a bit more deeply and look for a service with three key characteristics: authoritative data sources, scientific forecasting models, and a comprehensive range and quantity of data available for viewing. And if the service also had a visually pleasing design, I certainly wouldn’t mind.

It took several years, and I used various more or less known and utilized services, but today, I can confidently say that all my “weather deciphering” activities (because this still isn’t about planning) are conducted on Windy.com.

What is Windy.com?

Windy.com is, first and foremost, a portal. Launched in 2014 as a simple visualizer for wind direction and intensity information, accessible on its namesake website, it has evolved over the years into a full-fledged professional platform for meteorological analysis. It is completely free (we’ll discuss the paid Premium mode later) and offers a free smartphone and tablet application (which I highly recommend you download right away).

NOTE: The images in this article were all created using the tablet application. This allows for images that are well-suited for both web browsing and viewing on smartphones (or tablets, of course). Everything I will discuss and show you can be viewed on both the web and smartphones in exactly the same way.

The di Windy.com Interface

Like any good weather site or program, Windy is essentially divided into two areas: a map and a menu with what we’ll call selectable layers, which are data points of interest that can be graphically displayed.

Starting with the map, depending on the active layer, you’ll see various color-coded representations that are quite intuitive. If it’s your first time accessing Windy.com or if you haven’t registered (I recommend doing so since it’s free), you’ll see the wind map displayed. While useful for certain purposes, it’s not the most important one for us, so I suggest setting your profile to open with the “Clouds” view.

The menu gives you access to favorites, settings, and more (we’ll explore some cool features later)… Most importantly, as I mentioned, it gives you access to the layers. Windy.com offers over 50 layers, so you can imagine the potential.

I’ll now list the layers I use most frequently (which I’ve set as favorites in the app for quick access). We’ll discuss each layer individually before considering some strategies for combining the various pieces of information.

The layers I use most frequently (roughly in order of usage) are as follows:

• Clouds
• High Clouds
• Medium Clouds
• Low Clouds
• Satellite
• Weather Radar and Radar +
• Rain Accumulation
• New Snow
• Waves
• Wind and Wind Gust

You can activate one layer at a time, and you can display additional information that can be turned on and off from the menu just below the layer list. The isolines and webcams are probably the most useful of these, but we’ll discuss those later.

Let’s now examine the individual layers I use and the information I try to gather by using them, either individually or in combination.

The “Clouds” Layer

I’m starting with this layer because I believe it’s the most important one for a landscape photographer, so we’ll spend some time on it.

As the name suggests, the Clouds layer represents the predicted cloud cover.

If you’re used to other weather applications, the map visualization might not be particularly intuitive at first, or it may take some time to get accustomed to (at least that was the case for me).

That said, the visual information is very straightforward: areas with yellow indicate areas with NO clouds. Clouds are shown in shades of grey, ranging from grey (a few clouds) to white (many clouds).

If rain is expected, it will appear in shades of blue, from deep blue (light rain) to light blue (heavy rain).

This is why, honestly, I believe it’s not particularly intuitive, perhaps also due to the barely visible legend (located under the time slider in mobile versions and beneath the weather model selector in the desktop version). That said, it is possible to customize the map’s colors, which I experimented with for quite some time, but I strongly advise against this. You’ll get used to it, believe me.

I mention “cumulative coverage” and “forecast” because this visualization doesn’t provide a real-time representation of cloud cover nor an instantaneous forecast for a specific time, which I think is the most challenging part to grasp.

In fact, the display you see, regardless of the day and time selected with the appropriate slider, represents a forecast of cumulative cloud cover over a certain period and with a specific resolution.

The weather model being used determines this period and resolution.

For example, we might have a weather model with a 3-hour cumulative period and a 5 km resolution.

The resolution is easy to understand: 5 km means that the most detailed information the model can give me is over an area of 5 km. The higher the resolution, the better. For instance, a model with a 2 km resolution would allow for different weather forecasts for two neighbouring villages, whereas a 5 km model would not.

The period, however, is a bit more challenging to digest. A 3-hour cumulative forecast, as in our example, refers to the amount of clouds accumulated or passed over a specific area during the last 3 hours relative to the selected reference time.

This type of visualization can be helpful in understanding not only the current state of cloud cover but also for getting an idea of how it has developed over a short period. Essentially, you’re not just looking at the current situation in a single time frame but also considering the recent evolution of clouds, offering a more dynamic and complete perspective of the meteorological phenomenon.

There are various weather models, and depending on the portion of the map we are viewing, Windy allows us to choose our preferred model through the selection menu. Once the model is selected, we can also access its main details, such as the accumulation time, resolution, update frequency, who created it, and for what purpose.

Caution! A model is not necessarily better just because it has a higher resolution and a shorter accumulation period. I could create a “Frank” model with a 1-hour accumulation period and a 50-meter resolution, but if the model is poorly made (as it would be if I designed it), you would surely get information that is as precise as it is wrong! So when selecting a model, please don’t base your choice (only) on accumulation periods and resolution but also on the quality of the model itself for the information you seek. The answer to “Okay Frank, so which model should I use?” will come soon, so be patient.

Good to know: Beyond the graphical visualization, you can access numerical data by tapping on a specific point on the map. With a first tap, you’ll see the cumulative cloud cover percentage for that specific location, and by pressing “Forecast for this location,” you’ll get a more detailed hourly breakdown of the forecast.

If the detailed view doesn’t seem to offer better information than the graphical version on its own, let me teach you a little trick. At the bottom of the detailed view (whether you’re using the web or app version), you’ll notice that the display mode is set to “basic.” Instead of this view, tap on “Meteogram“: as you’ll see, you can now access a hybrid view with cloud cover divided by cloud altitude!

…and this leads us directly to the next layers!

The Clouds Layers (High, Medium, Low)

We’ve seen that with the Clouds layer, we can view the overall cloud cover, which doesn’t account for the three-dimensional nature of our skies.

As skilled photographers, however, we know that not all clouds are the same and, more importantly, that clouds at different altitudes behave differently, both in terms of their potential for coloring during sunrise and sunset and their movement.

This already long article can’t delve into the details of what to expect from each type of cloud (something you can learn in my field courses, of course 😉 ), but it will help you understand what is forecasted for the three atmospheric layers where clouds are typically divided: High, Medium, and Low.

By activating the respective layers (High Clouds, Medium Clouds, Low Clouds), you can view the exact same information as the Clouds layer (overall cloud cover) but for each specific cloud layer.

Windy divides clouds into these three bands:

• Low: from the surface to 2,000m (6,500 ft)
• Medium: from 2,000m (6,500 ft) to 6,000m (20,000 ft)
• High: above 6,000m (20,000 ft)

This classification is a general convention and can vary slightly depending on atmospheric conditions and the model’s specific interpretation of meteorological data, but it’s more than sufficient for our needs.

In photography, knowing the cloud cover forecast by altitude can really make a difference.

For example, imagine a situation where only high clouds are present: the Clouds layer wouldn’t give us this information (unless we use the Meteorogram), but thanks to the High Clouds layer, we could immediately understand that the right time to shoot might be just before sunrise or after sunset.

Similarly, an extensive layer of medium clouds seen in the Medium Clouds layer might indicate that, regardless of the other layers present, the light could be blocked at the horizon, suggesting we should seek a more favorable spot.

Finally, a layer rich in low clouds, visualized through Low Clouds, could guide us to start chasing light a few hours before sunset and focus on side-lit compositions.

And let’s not even get started on combining clouds in different layers, which almost infinitely multiplies the possibilities!

Again, it’s impossible to discuss detailed scenarios in this article, but I’m sure you’ve understood that Windy.com provides us with all the tools to maximize our chances of being in the right place at the right time, even with just the four cloud cover layers.

(But Which Weather Model Should I Use?)

As you’ve seen, Windy.com offers several forecasting models, and typically, each provides slightly different information.

We’ve already understood that choosing the model that gives us the desired result isn’t a good approach. The next step is understanding that models are often developed for specific needs and geographic areas. For example, the NEMS model is designed to be particularly accurate in the Alpine region, characterized by its mountainous terrain, while the UKV model is more precise in the UK area, where Atlantic influences are predominant.

For our purposes, I believe there are three key aspects to consider when choosing a model:

1. Geographic Area: A model developed for a specific area is generally more accurate.
2. Model Resolution: The higher the resolution, the more detailed information we can leverage.
3. Update Frequency: More frequent updates allow for more accurate decisions.

Getting to the point and considering that I live in Europe, where my experience is primarily with European models, although Windy.com actively promotes the ECMWF 9km (European Center Medium Weather Forecasts 9km), I’ve recently found that the most accurate information on cloud cover often comes from using the ICON model (D2 2.2km if available or EU 7km). This model offers higher resolution (2-7km versus 9km for ECMWF) and has a shorter forecast update interval (for every ECMWF update, you get at least two ICON-EU updates). I only use the UKV 2km model in the UK.

In the USA, I found the HRRR CONUS 3km model to be very reliable. Still, the territory is vast, and my experience was only for a few weeks, so I can’t give you a definitive opinion. As for the rest of the world, my knowledge is unfortunately quite limited.

But regardless of my experience (which might not be geographically relevant for everyone), how can you determine if a model provides reliable forecasts?

This question brings us to the next layer 🙂

The Satellite Layer

As the name suggests, this layer allows me to view an image of the current cloud cover provided by geostationary satellites.

This visualization is extremely useful because it’s no longer a forecast but a real-time snapshot of the actual situation. It enables us to see the real cloud cover over our heads while heading out to the field, for example.

So, while it’s useful for making the final decision on a shooting spot and moving towards the sunset location, it’s doubly useful at dawn when everything is dark, and looking out the window isn’t enough to decide whether it’s worth staying in bed or getting up to shoot.

Finally, this layer allows us to approximately validate the accuracy of cloud cover models.

By keeping in mind the fundamental difference between the Satellite layer, which provides a real-time visualization, and the Clouds layers, which offer cumulative forecasts, we can compare the instant view from the Satellite layer with the forecast from the chosen model at the same hour to gauge how accurately the model is predicting.

If the real-time visualization from the Satellite layer closely matches what a Clouds model shows for that same hour—considering that the forecast displayed by the Clouds layer was made several hours earlier (as indicated by the model details)—we can assume that the forecast is reliable or at least performing well.

This isn’t an absolute truth (because, for example, very fast-moving weather systems could skew the concept of a three-hour cumulative forecast), but it has indeed saved me on several occasions.

Let’s break down my writing with a simple, practical example if it seems too complicated.

If we compare the Satellite layer for a given area with the forecast made by the NEMS model two hours ago and see that the current conditions don’t match the forecast well, it could indicate that this forecast is too old to be considered reliable at this moment, or it might suggest that the model isn’t accurate for this area (in this case, likely the latter, since we know NEMS works well in the Alpine region but not so much in the UK).

On the other hand, if we compare it with the UKV model (which updated its data an hour ago), we might see that the situation visually matches the Satellite layer much more closely. This would lead me to consider the UKV model more accurate and reliable as a forecasting model for the decisions I need to make right now.

It’s clear that the ideal situation is to have our preferred model recently updated and closely match what the Satellite layer shows, but this isn’t always the case, and sticking to inaccurate information can only harm our photography outing.

Adaptability is always the key to success.

The Weather Radar and Radar+ Layers

The Weather Radar layer is a measure of reflection made with a Doppler radar. It shows current rain/snow/hail or the water accumulation within a cloud. In short, it provides an instant measure of whether a cloud is precipitating or is very close to precipitating.

I find this visualization very useful for various analyses.

First and foremost, it helps me understand where it’s actually raining and where it’s not before heading out, allowing me to identify where the weather disturbances are most intense. For instance, if it’s a widespread and continuous rain (like from nimbostratus clouds), I can quickly determine that it’s not even worth leaving the house. On the other hand, if I’m chasing storms (such as cumulonimbus clouds that are actively precipitating), it helps me pinpoint exactly where to go.

Additionally, knowing when a downpour is about to hit me while I’m already out shooting works very well. Although this layer can’t be used for forecasting (well, it can, but it’s not reliable), it allows us to look at the past to understand the trajectory of the rain.

Beyond rain, I’ve found the Weather Radar layer extremely useful during snowstorms in the Lofoten Islands, where storms often follow one after another. With this tool, I was able to see the extent and direction of the storms around me while staying dry in the car, only stepping out to shoot between snowfalls (and also determining if it was a quick storm or something more extensive, in which case it was better to give up). Incredibly useful!

Recently, Windy.com introduced a new layer called Radar+, which allows you to simultaneously view the Doppler radar for rain (essentially the Weather Radar layer) alongside the satellite image of the current cloud cover (the Satellite layer).

This combined version provides a comprehensive view of the real-time situation (not a forecast), showing both the overall cloud cover and where it’s raining. It’s perfect for conducting all the analyses and considerations we discussed earlier, but now in one unified view!

The Rain Accumulation and New Snow Layers

One layer I use quite often, especially when I’m in particularly rainy areas (like Scotland, for example), is the Rain Accumulation layer.

This layer provides a very intuitive view of rain accumulation over the coming hours and days.

In my opinion, this tool’s greatest utility lies in its ability to help us anticipate the arrival of strong or stationary weather systems, which might prompt us to adjust our plans for the next few days slightly (or completely).

I use this cumulative view because this information isn’t easily obtained from the other layers we’ve discussed (particularly the Clouds layer). Those provide either instantaneous forecasts or cumulative views limited to a maximum of three hours. In contrast, the Rain Accumulation layer allows us to assess the situation after several days of accumulation.

The New Snow layer is a “variant” of this layer that I use when travelling in winter. Despite the name, it also shows accumulations of fresh snow, ranging from 12 hours to 10 days.

It’s incredibly useful, especially when visiting places like the Lofoten Islands or Iceland, where snow and rain (which can wash away the snow) often alternate. This layer helps me anticipate when I can expect to find beaches covered in fresh snow!

The Waves Layer

As a seascape photographer, I can’t overlook the Waves layer, which visualises waves height and period (considered as a combination of swell and wind waves).

To put it in more photography-friendly terms, this layer tells us whether we’ll be facing rough seas, which can be useful for several reasons.

Firstly, if you’re a storm chaser, this layer lets you see when the storm surge is expected to reach the coast and what kind of waves to expect (although it’s important to remember that wave height will be influenced by the seabed near the shore, which the various models obviously can’t take into account).

As a landscape photographer, this layer helps you make decisions about your shooting location: if rough seas are expected, I would definitely avoid low spots like beaches or even potentially dangerous areas like low cliffs.

For long exposure photographers, this layer is even more valuable because if the sea is too rough, even shooting from a high vantage point might result in too much “whitewash,” meaning that the long exposure could result in a sea that appears too white due to waves crashing against the coast (which can sometimes be a nice effect, but not always). For this reason, if the sea is too rough, I would choose a spot where the seabed drops off steeply near the shore, to minimize the effect of the whitewash from the waves.

Once again, Windy.com is an invaluable tool for helping me make informed decisions before heading out to the field.

The Wind and Wind Gusts Layers

Given that the application is named after the wind, it’s impossible not to mention the layers related to it! Jokes aside, information about wind direction and intensity is incredibly valuable for us.

Regarding wind direction, you may have noticed that it’s possible to visualize it in almost all the other layers (though you can remove it if you prefer). I believe this is precious because, for instance, it can give us an immediate idea of where the wind is pushing the clouds.

However, knowing the wind’s intensity is equally important, as this knowledge is crucial both for our safety and the success of our photography outing.

A very practical example comes from a recently concluded workshop in the Outer Hebrides, Scotland, while I was writing this article. In this region, I typically group the photography spots into three geographic areas, and depending on the weather, I finalize the outing in one of them. When storms arrive, these islands, exposed to the Atlantic Ocean, are also accompanied by strong winds, making it potentially disastrous to head to a location even if the clouds and sea appear perfect.

That’s why I use the Wind layer, but even more so the Wind Gusts layer, which indicates the expected wind gusts in a specific location. While wind itself can often be managed, wind gusts are a nightmare, as they create sudden bursts of pressure on the tripod, which could compromise the sharpness of the image—especially if you’re facing a headwind and can’t shield the camera with your body.

In short, if everything is perfect in a location but 70 km/h frontal gusts are forecasted, it’s probably best to find another spot!

The “Missing” Layers on Windy.com

As comprehensive as Windy.com is, there are certainly a few pieces of information that would be incredibly useful for us photographers.

For me, the most noticeable omission is related to tides.

Yes, it’s true, there is a sort of tide layer on Windy.com, which can be activated when requesting more information about a selected point on the map, but it’s quite rudimentary, with few available buoys and limited usefulness for us. For this reason, I still rely on tools like Nautide or Tidal Planner for tide information, though these are only free for daily forecasts. I believe this gap exists because reliable tide models are few, and their licenses are regional and very expensive, which doesn’t align well with Windy.com’s philosophy.

Another missing layer, which I don’t personally miss (though I mention it because I’m often asked about it), is one for forecasting cloud colorization.

As many of you might know, some services and apps promise to forecast the likelihood of clouds coloring during sunrise and sunset. The idea is good, but unfortunately, we are still far from obtaining even remotely reliable results. This is because predicting such probabilities requires interpolating data from different sources, accounting for many variables. The more variables involved, the lower the chances of accuracy, especially since most of these variables are themselves independent forecast models (cloud type, cloud base height, visibility, wind, location, apparent horizon issues for each cloud layer…).

So, don’t feel like you’re missing out on Windy.com for not having layers like the often disappointing Skyfire, SunsetWX, or the even worse Viewfindr. If you see miraculous forecasts and photos of epic successes, remember that even a broken clock is right twice a day, and you don’t need an app to figure out whether a single layer of clouds will color or not.

That said, I hope (and believe) that we will eventually reach reliable forecasting models in this area, thanks to technological advancements both in data computation and in our understanding of atmospheric phenomena (after all, remember how unreliable weather forecasts were just 10 years ago?!).

While it may be unrealistic today to expect reliable colorization forecasts, I do believe the technology exists to provide us with more information, such as answering the question we landscape photographers often ask: “Will those distant clouds block the light just at sunset?” (Note: this is very different from asking if the clouds will color).

Unfortunately, as of today, Windy.com does not directly provide a layer to answer this question. However, since (almost) all the necessary data is available, I’ve created a (free) plugin for Windy.com. This plugin doesn’t give an absolute answer but at least provides an indication of whether clouds at the point where the sun rises or sets might intersect the horizon due to the Earth’s curvature, potentially blocking the light (curse this non-flat Earth!).

It’s called “Clouds Horizon Distance” and you can find it HERE.

Additional Features of Windy.com

As I mentioned earlier in this lengthy article, beyond all these layers, Windy.com has a few hidden features that are really interesting.

Among these, I’ll highlight three.

The first is the ability to upload and view your own points of interest. By going into the menu, you’ll find a function called “Display KML,” which, as the name suggests, allows you to upload coordinate files in KML format (and not just KML, actually). The great benefit for me is that when I create a plan using PhotoPills, for example, I can export the file and load it onto Windy.com. This allows me to view all my photography spots on the weather map and have a super accurate view of the weather conditions at each shooting location in real time!

As you may have gathered from what I wrote earlier, the second is the ability to install plugins. The Windy.com community is a fantastic meeting point for developers and many enthusiasts who can contribute to the application. By going into the main menu and selecting “Install Windy Plugin,” you’ll find a list of third-party plugins that effectively add additional layers. These plugins are very specific, as they cover niche needs that aren’t addressed by the standard layers. If you want to try creating something yourself, you can start HERE

The last feature I recommend taking advantage of is the Webcam function. Windy.com allows you to view the location and images of various webcams on any layer. As we discussed earlier, especially cloud models use a 3-hour average, so the information should be taken with a grain of salt. Besides comparing with the satellite image, as we mentioned before, it’s always useful to check the real-time situation using the available webcams to see how accurate the model is. When I’m travelling, just before heading out to the planned location, I always do a final check on the webcams to decide whether to stick to the plan or make a last-minute adjustment. Believe it or not, several photos that I consider successful planning outcomes were born from this process.

Windy.com Premium Subscription: Is It Worth It?

As I’ve mentioned several times, Windy.com is completely free, but it also offers a paid Premium subscription (as of the time I’m writing this, it costs €21 per year).

Is it essential? Absolutely not. Would I recommend it? Absolutely yes.

With Windy Premium, you won’t gain access to new layers or features (which is why it’s not essential), but you will benefit from more frequent data updates, leading to more accurate forecasts.

In practice, while the weather models we discussed earlier provide updates every 12 hours in the free version, with the Premium version, they update approximately every 6 hours. You can imagine how valuable this is for us.

Additionally, you’ll have access to hourly forecasts instead of every 3 hours (note that the visualizations remain accumulated over 3 hours), the ability to loop the Radar and Satellite for 12 hours, historical data up to a year for both Radar and Satellite, and much more.

Again, it’s not indispensable, but if you want to fully maximize the potential of all the tools we’ve covered, the mere fact of going from 2 updates per day to 4 makes the subscription well worth it.

Conclusions

While having complete control over the weather during our photography outings remains a dream (but then again, where would the thrill be for those who love the experience?), being able to make informed decisions based on accurate and well-organized information has never been easier thanks to Windy.com.

With an array of visualization modes and selectable weather models, Windy.com provides us with an additional ally in our photography gear: the knowledge of what’s happening around us, enabling us to maximize our chances of using the weather to our advantage during every outing.

Completely free, it works on both laptops and mobile devices, and with a small annual subscription, it offers more frequent data updates and thus more accurate forecasts.

The sheer amount of information available might seem intimidating, but I hope this article has helped you take your first steps.

If you want to explore its usage more deeply, I’d be happy to help. There’s no better opportunity than joining one of my field workshops or booking a private online lesson!