How much snow falls on the world’s mountains each year?

January 26, 2021 | Eric Verbeten

The annual cycle of snowfall and snowmelt over the world’s mountains provides essential freshwater for ecosystems and the billions of people who rely on that water to live. Understanding this cycle is becoming more urgent as Earth’s climate changes.

The advent of advanced satellite instruments has made it possible to gather consistent data on snowfall from a consistent observing platform. Using these satellite data and comparing them to numerical weather models, scientists at the University of Wisconsin–Madison Space Science and Engineering Center, Department of Atmospheric and Oceanic Sciences, the Center for Climatic Research, NOAA and NASA found that 4 to 5 percent of global annual snowfall occurs in mountainous regions each year. The results of their research were published in the journal Cryosphere.

Anne Sophie Daloz is a scientist with CICERO in Oslo, Norway and researches ways to better predict annual snowfall in mountainous regions around the world. Credit: Anne Sophie Daloz

“Snow is a crucial component when working on these questions, we really need to better understand our ability to diagnose regional trends in snowfall and what it means for the snowpack,” says Anne Sophie Daloz, lead author and a researcher with the Center for International Climate Research in Oslo, Norway and former SSEC scientist. “This will contribute to better estimates of water availability at the regional level which is crucial for people’s livelihood in many regions of the world.”

Historically, scientists have relied on a network of surface instruments to provide measurements of snowfall. These networks; however, can be inconsistent, typically located near population centers, and unavailable in mountains or over the oceans. The result is a dataset that may not accurately represent actual global snowfall totals.

Through a method known as reanalysis, Daloz and her team compared actual snowfall observations captured by the NASA CloudSat weather satellite (from 2007-2016) to various global gridded datasets. A reanalysis is based on a numerical weather model incorporating millions of observations. Because different models have different strengths, the team compared the satellite results to results from five separate reanalyses. Doing so allowed them to identify biases in each dataset and how those biases might influence outcomes.

The scale of the study is massive. For that reason, Daloz and her team partitioned the globe into four regions including North America, South America, Eurasia and Australia. Reanalysis relies on multiple variables (like terrain, temperature and humidity to name a few) to forecast and evaluate climate behaviors. Small changes in these variables can result in big changes in precipitation over time, which is why Daloz investigated how well each model matches its counterparts. Due to this high-degree of variability, certain models perform better or worse at predicting snowfall, depending on the structure of the algorithm.

“In South America for example, there were large differences between data sets,” says Daloz. “If you translate that in terms of water supply, that makes a huge difference.”


The Cryosphere paper investigates total annual snowfall over the world’s mountainous regions. For this study, the researchers use a broad interpretation to identify mountains in each of the four study regions.

Melissa Wrzesien, a hydrogeologist with the NASA Goddard Earth Sciences Technology and Research, and co-author, says the team uses a mixture of elevation, slope and local relief to classify mountains.

“An area that maybe doesn’t have a large elevation but has steep slopes around it could be considered a mountain. Similarly, an area that stands out from the land around it – which means it has a large local relief, or elevation difference – like a plateau or mesa, could be considered a mountain.”

In total, the 4 to 5 percent of annual global snowfall identified in the paper is equivalent to 1773 km3 (425 mi3) – enough to cover the state of Wisconsin in more than 10 meters (32 feet) of snow. If converted to precipitation in the form of rain, Wrzesien says it would submerge the entire state under a meter of water.

Its distribution around the world; however, is crucial to the survival and way of life for billions of people in and around these mountainous regions.

“Accurate measurements of snowfall in mountain regions is incredibly important for understanding our water resources,” says Wrzesien. “Snow that accumulates in mountain regions later melts into nearby rivers. Then the water is often used by people, whether for generating hydroelectric power, irrigating crops or being used as drinking water.”

Because of these differences, Daloz is shifting research from a global scale, to a more focused regional approach. This shift will allow Daloz and her collaborators to take advantage of satellite data with higher resolutions (and greater accuracy) available on other satellites.

“If we want to improve our ability to observe or simulate this component, I think we need to go to the regional level for getting a better understanding of the processes leading to the different biases we observe in the paper,” says Daloz. “In our work, we only included satellite observations and reanalyses, but the next step would be to work with regional climate models. [These] can reach very high spatial resolutions of less than 3 km (2 miles).”

Daloz hopes that piecing together the regional snowfall variations into a global picture of snowfall processes will increase understanding and provide crucial information for planners tasked with anticipating water shortages – or abundances – in an effort to meet growing food and energy needs.

This work was supported by NASA and the Center for Climatic Research. Photo credits: Anne Sophie Daloz