# Orographic lift: why one side of a range gets more snow > Why the windward side of a range gets more snow, how much less the lee usually gets, and how to use storm direction to choose a mountain. Source: https://skibudsapp.com/guides/orographic-lift-and-snow Updated: 2026-10-24 Orographic lift is what happens when wind pushes moist air up a mountain range. The air cools as it climbs, its moisture condenses, and most of the snow falls on the slopes facing the wind. The far side, the lee, gets less, and how much less is the useful number: in a typical Front Range upslope storm, at most 20 to 40% of the biggest windward total falls on the lee side, according to [a study of Colorado's March 2003 storm](https://www.weather.gov/media/gjt/Research/Exceptional_Mesoscale_2003.pdf) by a team that included National Weather Service and NOAA meteorologists. So pick the mountain the storm is blowing toward. ## On big ranges, the snow peaks on the windward slopes before the crest Gerard Roe, a University of Washington atmospheric scientist, reviewed the science of mountain precipitation for the [Annual Review of Earth and Planetary Sciences](https://earthweb.ess.washington.edu/roe/GerardWeb/Publications_files/Roe_OrogPrec_AnnRev05.pdf) in 2005. He calls a rain shadow across a range that sits crosswise to the prevailing wind "one of the most confident expectations in atmospheric science." His best example is New Zealand's Southern Alps, which face a steady line of storms off the open ocean. The western lowlands get 2 to 3 meters of precipitation a year. On the steep windward slopes that climbs to 11 to 12 meters, about 36 to 39 feet of water, roughly 20 kilometers before the divide. Across the crest, the eastern plains get less than 1 meter. Two details from the review matter for skiers. On big ranges the heaviest precipitation usually falls on the windward slopes before the crest, while on smaller hills it peaks nearer the top. And because snow takes time to form and fall, the wind carries some of it past the crest, so the first few miles of the lee can still do well. The effect is strongest when moist air blows hard and straight up the slopes; Roe notes the heaviest rates in a storm come during the hours when that's true. Farther down the lee, the air sinks and warms, clouds thin out and snowfall drops off. A strongly dry lee is called a rain shadow. When that sinking air is also warm and windy, it can become a Chinook, which [Chinook winds and snow](https://skibudsapp.com/guides/chinook-winds-and-snow) covers. ## In March 2003, an east wind flipped Colorado's usual pattern Colorado's prevailing winds come from the west, and the [Colorado Climate Center](https://climate.colostate.edu/climate_long.html) describes the result: Pacific storms moving east drop much of their moisture on the mountaintops and west-facing slopes, leaving the eastern slope relatively little, especially in midwinter. From March 17 to 20, 2003, a slow-moving low over the Southwest did the opposite, driving moist easterly winds from the Gulf of Mexico up the Front Range. The windward side fit Roe's pattern. The [National Weather Service in Boulder](https://www.weather.gov/media/bou/March03Snowstorm.pdf) recorded 87.5 inches atop Fritz Peak and in Rollinsville in the foothills, 55 inches at [Eldora](https://skibudsapp.com/resorts/eldora), closer to the crest, and 31.8 inches in Denver, the city's biggest single storm since 1913. The lee side didn't. By the study's own typical figure, the slopes west of the Divide should have seen a fraction of that. Instead about 2 meters, roughly 90% of the windward maximum, fell in places on the lee side, with record snowfall at [Winter Park](https://skibudsapp.com/resorts/winter-park), about 7 kilometers past the crest, where snow fell at times at 1 to 3 inches an hour. The authors found easterly flow reaching above the 250-millibar level, more than 30,000 feet up, and air at 80% humidity or more up to about 20,000 feet over the lee. The snow there was also much lighter, around 6% water content against 10 to 16% at lower elevations on the Front Range. The guide to [snow-to-liquid ratio](https://skibudsapp.com/guides/snow-to-liquid-ratio) explains how water content relates to snow density. The study's reach is narrower than its headline ceiling suggests. It documents one storm on one range in detail: the totals, the snow density, the forecasts. The 20 to 40% ceiling is stated as the typical behavior of Front Range upslope storms, not measured across many storms in the paper, and it describes easterly storms, the reverse of Colorado's prevailing westerly pattern. It's a useful number for how much the lee loses, and a rough one. ## Models miss the amounts but get the pattern Orographic snow depends on terrain details that weather models smooth over. In 2003, the Eta model, one of the government forecast models of the time, predicted about 1 meter for Winter Park, roughly half what fell. Roe's review has a similar case in Washington's Olympic Mountains, where a high-resolution model underestimated one storm's total by about a factor of two but got the ratio of ridge to valley precipitation right. That pattern is the useful part. Roe's summary is that amounts for any single storm are hard to predict, while the patterns, even at the scale of a few kilometers, hold up much better. Trust a forecast more on which side of the range wins than on the exact number. [Why snow forecasts disagree](https://skibudsapp.com/guides/why-snow-forecasts-disagree) explains the rest. ## SkiBuds will leave terrain corrections out until they're tested The SkiBuds forecast engine contains a terrain-alignment calculation that compares wind direction with the direction a slope faces. It doesn't use it to change snowfall totals. In the current forecast the orographic adjustment is fixed at no change: the engine keeps each source's own amount for each elevation, because neither the wind at the surface nor an unverified reading from higher up proves how much more snow a windward slope will get. A correction like this can make a forecast look more precise while making it less accurate, and 2003 shows how far off a rule can be. SkiBuds will apply one only after it has been checked against observed snowfall. Until then, the base, mid-mountain and summit views in the rebuilt app, which is coming soon, will show the forecast sources' elevation amounts without an extra terrain multiplier. ## Ski the side the wind hits, and expect the lee to get less than half as much The rule: ski the side the wind hits, assume the far side gets less than half as much, and let the morning reports prove you wrong. To apply it: 1. **Find the wind direction aloft, not at the base.** National Weather Service forecast discussions often describe the flow at 700 millibars, about 10,000 feet, which is closer to what a ridgeline feels than the wind in a valley. 2. **Favor mountains that face that wind,** and ones on or just past the crest, where spillover lands. 3. **Discount the far side.** Plan on the lee getting no more than a fifth to two-fifths of the biggest windward total, unless the storm is unusually deep and moist. That ceiling comes from Colorado's upslope storms, so treat it as a rough guide on other ranges. 4. **Check observations after the storm.** Measured snowfall and the resort's report settle which side did best; see [snow depth vs snowfall vs SWE](https://skibudsapp.com/guides/snow-depth-vs-snowfall). [How to read a ski resort snow forecast](https://skibudsapp.com/guides/how-to-read-snow-forecasts) explains base, mid-mountain and summit forecasts, and the [mountain region guides](https://skibudsapp.com/mountains) help you compare nearby resorts. The one-line rule has a cost, and 2003 shows it: it would have sent a Denver skier to Eldora's 55 inches while Winter Park, just over the Divide, got close to 80. So when the forecast for the Front Range calls for deep, moist easterly flow, step 2 is the one to lean on: look at the mountains just west of the crest too, before you pick one.