Lumavik against the other aurora apps, honestly.

We are not the biggest app in this category and it would be silly to pretend otherwise — the ratings column below makes that obvious at a glance. What these pages try to do instead is answer the question a comparison page usually dodges: given how you actually use an aurora app, which one should you install?

Every page in this section names at least two situations where the other app is the better choice, and means them. That is not modesty. A comparison that concludes we win on every axis is worthless to a reader and transparently self-serving, and you would stop trusting the rest of the site the moment you noticed.

The thing worth knowing before you compare anything

Every aurora app on both stores reads the same free NOAA feeds. There is no proprietary aurora forecast, there is no secret model, and any app implying otherwise is describing marketing rather than physics. The Kp index, the three-day outlook, the OVATION oval and the L1 solar wind measurements are public domain and identical for everyone.

So the data is never the differentiator. What differs is everything that happens afterwards. Does the app score conditions against your geomagnetic latitude, or does it show you a planetary average dressed up as a local forecast? Can you set the alert threshold, or does it fire on somebody else's idea of exciting? Does it check whether it is dark where you are before waking you? Does it tell you when its own data has gone stale? Does it say no?

Those are the questions these pages are built around, along with the ones that decide real installs and have nothing to do with forecasting: how heavy the app is, how aggressive the advertising is, whether the maps are usable on a phone in the cold, and whether the thing is still being maintained.

How we source the facts

Ratings, rating counts, prices, sizes and update dates come from Apple's public App Store lookup API, captured on 31 July 2026 for the United States storefront. Feature descriptions come from each developer's own store listing. We do not speculate about features we cannot verify from those two sources, and we do not repeat complaints from reviews as though they were facts.

Two honest caveats about the numbers in the table. A rating count measures install base and age, not quality — an app with 68,000 ratings has been downloaded a very large number of times over many years, which tells you it works and that a lot of people found it, and tells you almost nothing about whether it suits you. And a recent update date is a better health signal than either the rating or the count, because an unmaintained app in a category that depends on live external feeds eventually breaks.

All of this changes. If a page here is out of date or unfair, including if you are the developer of one of these apps, tell us and we will correct it.

How aurora apps make money, and why it matters

Nearly every app in this category is free to download, which means the money comes from somewhere else, and where it comes from shapes the product in ways worth understanding before you install anything.

The largest apps here are advertising businesses. Free download, a large install base built over many years, banner and interstitial advertising against it, and an optional paid tier to remove the ads. That model rewards frequent opening, which is a mild but real incentive to make the forecast look encouraging. It also funds long-term maintenance properly, which in a category dependent on external feeds is worth a great deal — an app that is still being updated after six years has usually earned that.

A second group are subscription apps with a thin free tier. The incentive there points the other way: they need you to convert, so the free tier is often deliberately limited, and the honest question to ask is whether the paid features are things you will actually use or a list assembled to justify a price. We are in this group, with the free tier deliberately drawn to include the entire forecast rather than a teaser of it, and you should weigh that claim knowing we are the ones making it.

A third group are non-commercial or institutional: research groups and enthusiast projects publishing tools because the data is public and they care about it. These are frequently the most technically accurate things in the category and frequently the least pleasant to use on a phone at midnight, which is a genuine trade-off rather than a criticism.

And a fourth group are utility-farm apps, published in volume by studios that ship dozens of similar single-purpose apps across unrelated subjects. Some are perfectly decent. What they rarely have is anyone who cares specifically about aurora, which shows up in the details — the alert that ignores daylight, the score that never varies by location, the map that has not been updated since launch.

None of these models is disqualifying and we are not suggesting you avoid any of them. But knowing which one you are looking at explains most of what you will find when you open the app, and it is more predictive than a star rating.

A checklist you can use on any aurora app

You do not need our comparison pages to evaluate one of these apps. Install it and check seven things, which take about ten minutes between them and will tell you more than any review.

First, find out whether the score it shows you is local. Change your saved location from your home town to somewhere three hundred kilometres further north and see whether the number moves. If it does not, the app is showing you a planetary Kp with a friendly label on it, and it will be wrong for you in both directions — optimistic when activity is on the far side of the planet, pessimistic when a modest Kp is enough at your latitude.

Second, look for a timestamp. Live space-weather feeds fail regularly, and an app that cannot tell you how old its data is will happily show you a number from six hours ago as though it were current. This is the most common quiet failure in the category.

Third, open the alert settings and see whether you can set the threshold. An alert that fires on somebody else's definition of interesting is an alert you will turn off within a fortnight, and once it is off you have lost the entire value of installing an aurora app in the first place.

Fourth, check whether darkness is part of the logic. If you can set an alert in June at seventy degrees north and the app cheerfully offers to notify you at 2pm, nothing in it knows about the sun.

Fifth, look at the oval map on a phone in one hand. Many of these maps are desktop graphics scaled down, and reading one outdoors in the cold with gloves on is a different task from reading it on a laptop.

Sixth, check the update date on the store listing before you install anything. This category depends entirely on external feeds that change format occasionally. An app that has not been touched in two years is on a countdown to breaking, however good its ratings look.

Seventh, and least obvious: see whether the app ever tells you no. Open it on a quiet night with Kp 1 and read what it says. An app that finds something encouraging to report on a night when nothing is going to happen has told you exactly how much to trust it on the night when something is.

One thing deliberately absent from that list is the star rating. Ratings in this category are dominated by a bias that has nothing to do with software quality: people who saw the aurora leave five stars, and people who went out and saw nothing leave one, and both outcomes were mostly decided by cloud cover. Read the recent one-star reviews of any aurora app and you will find a large share of them describing weather. That is also why we would rather you used the checklist than the average.

Running two apps is a reasonable answer

Nothing forces this into a single choice, and among people who chase seriously the common setup is two: one app whose alerts they trust enough to leave switched on overnight, and one source of raw data they open once the alert has woken them. Those are genuinely different jobs. An alert has to be conservative, because its cost is measured in wasted nights and lost sleep. A data view should be maximally detailed, because by the time you are reading it you have already decided to be awake.

Very few apps do both well, and an app optimising for one is usually making the other worse on purpose. If you find yourself irritated that the alerting app does not show you the interplanetary magnetic field, or that the detailed one buries the single number you wanted at 1am, that is the design working as intended rather than failing. The install cost is zero, so the sensible move is to keep both and let each do the job it was built for.