Is March a Good Time to Ski? The Deepest Half-Month of the Season
Published 27 Aug 2026
Across the 27 resorts we track, the deepest half-month of the entire winter is the second half of March, at a median mid-mountain depth of 145 cm. That is deeper than the first half of February (138 cm), deeper than the second half of February (139 cm), and deeper than any half-month in January. The assumption that March is the start of the decline is, on 15 winters of history, wrong.
The caveat is the whole story though, because that median hides a split down the middle of the table. March out-depths February at 13 of the 27 resorts and at none of the six whose mid-mountain sits below 2,000 m. March is not later winter for everyone. It is later winter for the high resorts and the glaciers, and it is early spring for everything else, in the same fortnight.
Is March snow really deeper than February snow?
On the site-wide median, yes, and the peak is later than most people would guess. Taking the median across all 27 resorts of each resort’s own median mid-mountain depth:
| Half-month | Median depth, 27 resorts |
|---|---|
| Second half of January | 134 cm |
| First half of February | 138 cm |
| Second half of February | 139 cm |
| First half of March | 142 cm |
| Second half of March | 145 cm |
| First half of April | 142 cm |
| Second half of April | 121 cm |
The curve is close to flat from early February to early April and then falls off a cliff. That flatness is the useful finding: across a nine-week window there is barely 7 cm of median difference, so anyone choosing February over March on depth grounds is optimising something that is not there.
Per resort the picture is more interesting. Comparing each resort’s deepest February half-month against its deepest March half-month, 13 of the 27 are deeper in March. Below are the ten largest gains and the five largest losses.
| Resort | Mid-mountain | Deepest Feb half | Deepest Mar half | Difference |
|---|---|---|---|---|
| Cortina d’Ampezzo | 2,470 m | 91 cm | 113 cm | +22 |
| Lech-Zuers | 2,000 m | 133 cm | 152 cm | +19 |
| Bormio | 2,000 m | 135 cm | 152 cm | +17 |
| Stubai Glacier | 2,900 m | 158 cm | 174 cm | +16 |
| Livigno | 2,300 m | 141 cm | 156 cm | +15 |
| Saas-Fee | 2,980 m | 202 cm | 214 cm | +12 |
| Hintertux | 2,660 m | 148 cm | 159 cm | +11 |
| St Anton | 2,085 m | 163 cm | 173 cm | +10 |
| Davos Klosters | 2,215 m | 145 cm | 152 cm | +7 |
| Ischgl | 2,320 m | 162 cm | 167 cm | +5 |
| Obertauern | 2,000 m | 120 cm | 111 cm | -9 |
| Val Gardena | 2,000 m | 63 cm | 54 cm | -9 |
| Mayrhofen | 1,565 m | 126 cm | 118 cm | -8 |
| Flachau | 1,450 m | 115 cm | 104 cm | -11 |
| Zell am See | 1,350 m | 123 cm | 107 cm | -16 |
The top of that table is not a random selection of resorts. It is the high ones, the glaciers, and the two Italian areas that sit where late-winter systems arrive. The bottom of it is the low Salzburgerland and Tyrolean valley resorts, plus Val Gardena and Obertauern, which are nominally at 2,000 m and behave like lower resorts for the same reason they do in every other guide on this site: not enough snow falls on them in the first place.
Cortina is the sharpest single case. The first half of March is its deepest half-month of the entire season at 113 cm, against 90 and 91 cm in the two halves of February, so roughly a quarter deeper than the month everyone books. That is the same resort that finishes 16th out of 22 in our snow-sure ranking. A resort can be unreliable across the season and still have March as its best window.
Where does the second half of March fall off?
Below 2,000 m, without exception, and the rule is unusually clean. Every one of the six resorts we track with a mid-mountain below 2,000 m loses depth between the first and second half of March, by 3 cm to 11 cm. Ten of the 21 resorts above 2,000 m gain depth over the same fortnight, five of them glaciers.
This is the point in the season where a single elevation number starts predicting your holiday again, the way it does in November and December and stops doing in January. The mechanism is the one covered in the Easter and late-season guide: by late March the sun is high enough that a low or south-facing slope loses more to melt each week than the storms put back, while 600 m higher the same storms are still pure accumulation.
What that means for booking is narrower than it sounds. It is not a warning off the second half of March. It is a warning that at a low resort, the second half of March is the last week the snow is still roughly what it was in February, and the first half of April is not.
Does March cost you reliability compared with February?
In the first half, nothing at all. In the second half, a little, and only at six resorts.
Reliability here is the site’s base statistic: the share of the last 15 winters in which a resort had a 30 cm or deeper base at mid-mountain by the checkpoint date, the 15th for a first half and the last day of the month for a second.
| Half-month | Resorts clearing 30 cm in 14+ of 15 winters |
|---|---|
| First half of February | 27 of 27 |
| Second half of February | 27 of 27 |
| First half of March | 27 of 27 |
| Second half of March | 21 of 27 |
| First half of April | 16 of 27 |
| Second half of April | 16 of 27 |
The first half of March is statistically indistinguishable from February. Every resort on the site cleared the threshold in at least 14 of the last 15 winters, exactly as in both February halves. There is no reliability argument for February over early March.
The second half of March costs six resorts their perfect record, and they are the predictable ones: Val Gardena drops to 10 of 15 winters, Madonna di Campiglio to 12, and Cortina, Saalbach-Hinterglemm, Kitzbuehel and Zell am See to 13. Every other resort is still at 14 or 15 of 15.
Then the first half of April takes five more resorts below that line, from 21 down to 16. That step is the actual boundary of the reliable season, and it is at the end of March rather than at the start of it.
Is March a dry month?
No, and this is the assumption that costs people the most. March has more snowfall days than February at 20 of the 27 resorts. The site-wide medians for whole-month average snowfall days are 13.4 in January, 11.2 in February, 11.4 in March and 9.8 in April.
So March is a wetter month than February and a drier one than January. It only reads as a dry month because the snow that falls in March is competing against sun and melt, so a given number of storm days converts into less standing depth than it would in January. That is also why the deepest half-month arrives at the end of March rather than in April: the snowfall is still arriving, and the melt has not yet won.
Four resorts get more snowfall days in March than in January, and three of them are the driest-January resorts on the whole site:
| Resort | Jan snowfall days | Mar snowfall days |
|---|---|---|
| Cortina d’Ampezzo | 7.6 | 9.0 |
| Madonna di Campiglio | 7.6 | 8.9 |
| Val Gardena | 7.0 | 7.5 |
| Obergurgl-Hochgurgl | 10.1 | 10.3 |
The three Italian ones are the same resorts our other guides single out for weak early-season numbers, and this is the other half of that pattern. The southern side of the Alps runs a later season than the northern side, so the months where the Dolomites disappoint and the months where they do not are not the same months. If you are booking a Dolomite resort at all, March is a better bet than January on the underlying weather, whatever the season-long ranking says.
Why March rather than April?
Because in April the lifts stop, and that is a harder constraint than the snow.
Of the 27 resorts, 26 are still open on 31 March. Only Seefeld in Tirol closes earlier, on 29 March. By 10 April 21 are open, by 20 April only 12, and by 30 April 10. The month where the season falls apart is April, not March, and it falls apart on the calendar rather than on the snowpack.
The snow follows the same shape. Comparing each resort’s second-half-of-April depth against its second-half-of-March depth, the retention splits by altitude harder than any other statistic on this page:
| Resort | Late March | Late April | Retained |
|---|---|---|---|
| Madonna di Campiglio | 90 cm | 27 cm | 30% |
| Val Gardena | 41 cm | 15 cm | 37% |
| Seefeld in Tirol | 92 cm | 36 cm | 39% |
| Zell am See | 97 cm | 42 cm | 43% |
| Kitzbuehel | 99 cm | 43 cm | 43% |
| Obertauern | 96 cm | 57 cm | 59% |
| St Anton | 173 cm | 155 cm | 90% |
| Stubai Glacier | 174 cm | 157 cm | 90% |
| Ischgl | 167 cm | 157 cm | 94% |
| Breuil-Cervinia | 164 cm | 177 cm | 108% |
A low resort keeps roughly a third of its late-March base into late April. A high one keeps nine tenths, and the highest keep all of it. So the honest framing of March is not “the last good month”. It is the last month in which the resort you pick barely matters, and April is the first in which it decides everything.
What the March numbers do not tell you
Depth is not the surface you ski on. By March a spring snowpack is on a freeze-thaw clock: firm in the morning, corn by mid-morning, heavy by afternoon on anything sunny. 145 cm of March base does not ski like 145 cm of February base, it skis better early and worse late. A March day is front-loaded in a way a February day is not, and the median depth statistic cannot see that at all.
Aspect starts to matter as much as altitude. In January a north-facing and a south-facing piste at the same elevation are close to interchangeable. In late March they are not. Our numbers are computed at a single mid-mountain point per resort, so they describe neither, and the difference between the two sides of a ridge in late March is larger than most of the gaps in the tables above.
Snowmaking is not modelled. The figures are natural snow only, so real coverage is often better than they suggest and never worse. This matters less in March than in December, because by March a well-run resort is mostly living off a base it built months ago rather than making new snow, and warm nights limit what the cannons can add anyway.
Closing dates move, and most of ours are estimates. Of the 25 resorts with a published close date, 18 are currently estimated from recent seasons rather than announced. Resorts also extend or cut the last fortnight on demand, not on snow. Treat the closing date on each resort page as the first filter and check it again closer to the time.
Easter can land inside March. Easter falls anywhere from 22 March to 25 April, and in an early-Easter year the second half of March is a peak-price fortnight rather than a quiet one. The Easter and late-season guide covers how much that moves the answer.
Fifteen winters is a deliberately short window. Peer-reviewed station analysis across the Alps (Matiu et al. 2021, covering 1971-2019) found the steepest snow-depth declines at lower elevations and in the shoulder seasons. The bottom rows of every table here are the ones most exposed to that trend, and a longer lookback would flatter them rather than describe them.
How should you book March?
- First half or second half? The first half of March is February with better daylight and no reliability penalty at any resort we track. The second half is where the altitude filter switches back on.
- If you are booking the second half, check the mid-mountain elevation. Above 2,000 m the base is typically still growing. Below it, every resort in our data is losing depth by then.
- Check the closing date before the snow statistics. March is nearly closure-free at 26 of 27 open on 31 March, but a late-March booking at a resort closing on 4 April leaves you no margin if you wanted to extend.
- On the southern side, prefer March to January outright. The Dolomite and Trentino resorts get more snowfall days in March than in January, and Cortina’s best half-month of the entire season is the first half of March.
- Read the resort page, not the month. By late March the weakest resort in our data is down to 41 cm against a 145 cm median, while in February no resort we track is below 60 cm. The month-level answer is good news; the resort-level answer is the one you are actually buying.
Then use the live side of the site in the week before you travel. From late autumn each resort page carries the current modelled and measured base, the 7-day forecast, the resort’s open status and the EAWS danger level for its region, refreshed every morning. In March that last one is worth reading with the clock in mind, because the spring wet-snow cycle makes the danger a function of the time of day as much as of the number.