Exploratory: generated automatically from the UmaTerra platforms' data; not independently validated or peer reviewed.
On 1 April 2026 the Okanagan’s snow stood at 58% of normal, the lowest since 1980, in a La Niña winter. This story reads the winter of Vernon and Kelowna, of the Silver Star and Big White mountains, and of Glacier National Park, which lies outside the basin, through the province’s snow bulletin, a reanalysis, stations and satellite; tests the idea that snow near 0 °C is what is lost first; and looks at an El Niño that may be the strongest ever measured, without trying to forecast it.
Loading the numbers…
measured by the bulletin, a station or a satellite. estimated a reading that depends on a choice (a grid, a window, a threshold). missing a part that did not respond this time.
Vernon and Kelowna lie on the floor of the Okanagan valley; Silver Star and Big White are the mountains above them; Glacier National Park lies east, in the Columbia Mountains, outside the basin. The story uses the three as a ruler: what snow and winter did where winter is mild, and where it is cold. For the basin’s hydrology, see Merritt and colleagues.
BC's River Forecast Centre sums up each basin's snow in an index: the percent of the 1991–2020 normal at snow-survey stations. On 1 April 2026 the Okanagan basin had the lowest value of any region, and the bulletin calls it the lowest since basin indices began in 1980.
Drag the bar: on the left, late March and April 2023, a winter with above-normal snow on 1 April; on the right, the same weeks of 2026, the lowest index. Sentinel-2, true colour.
The literature separates a snow drought from lack of precipitation from one caused by warmth, where the water falls as rain or melts early. The Okanagan index through the months shows that 2023 and 2024–2026 are different stories. In western North America mountain snowpack has been declining, and a low-to-no snow future is projected for parts of the region.
Hansen maps where the tree canopy disappeared each year, from fire, felling or pests: it does not say the cause. In red, 2023. The large scar west of Okanagan Lake lies where the West Kelowna story analyses the McDougall Creek fire; 2023 was also the year the Okanagan snow index collapsed from May to June (previous step).
In the ERA5-Land reanalysis the November–March mean temperature at Vernon, Kelowna, Silver Star and Big White has risen since 1981; the trend is the median slope across all pairs of winters (Theil–Sen), which is not led astray by one outlying winter.
Environment and Climate Change Canada's stations measure the same winter on the ground. They have gaps and moves, and a town's series are spliced without homogenisation; so the grid is the main series and the stations are the check.
AAFC’s crop map shows what the valley floor does with its land in 2025: orchards and vineyards, pasture, city. It is an agricultural inventory and a one-year snapshot: its forest and shrub classes changed between product versions, so no change is claimed. This story models neither irrigation nor runoff; for the basin’s hydrology, see Merritt and colleagues.
A freezing day is a day when even the maximum stays below 0 °C: snow that falls on those days stays. In the Okanagan's towns and mountains those days are becoming rare; at Glacier, they are not.
One explanation for the contrast: snow that falls near 0 °C is what is lost first. To test it, each of the 1,350 ERA5-Land cells of BC's southern interior was ordered by its 1981–2000 winter mean temperature and compared with its 1 April snow in 2012–2026.
MODIS has seen snow on the ground since 2000 and is independent of the reanalysis. Frequency is the share of cloud-free observations in which the pixel had snow in April; clouds hide some days, and snow days are cloudy, so it overstates snow, but the bias is the same every year.
NTEMS’s forest-age map, built from decades of Landsat imagery, gives each stand’s age as of 2019. The table splits each place by elevation band: how much forest there is where snow still sits and how old it is.
At Glacier the average winter is about 9 °C below zero; April snow covers the pixels nearly every year. But the Upper Columbia basin, where the park lies, also had weak winters: what the contrast shows is a difference of margin, not immunity.
The World Database on Protected Areas (WDPA) records Glacier National Park and the Okanagan’s protected areas, the same database Camposcope consults when a click lands outside any parcel.
NOAA's Climate Prediction Center now gives a high chance of a historic event, stronger than any El Niño since 1950. It now judges strength by the relative index (RONI), which runs below the classic ONI; the table shows the record and the three previous strong El Niños at the same point of the year.
Work on British Columbia links El Niño and the Pacific Decadal Oscillation to surface weather and to the duration of snow cover in the province, but a winter's response depends on more than the phase. The table shows what ERA5-Land says for each of the three strong-El-Niño winters since 1981, and for the 2025–26 La Niña winter.
The pattern expected from an extreme El Niño over North America, a warmer Canada and a wetter western US, rests on few events. Models project more frequent extreme El Niños in a warmer climate, and what happens to the teleconnections themselves is still uncertain. An event stronger than any on record has no analogue.
The NEX-GDDP-CMIP6 projection grid that muda-clima uses gives each 0.25° cell's annual warming between 1985–2014 and 2041–2070 under two scenarios, the median of three models. It does not hold winter separately: annual warming stands in for winter warming here, and the table says where each place would land on the test's curve. A reading of exposure, not a forecast.
Written with Claude Sonnet 5.5, by Anthropic, from the cited papers and bulletins and the numbers Earth Engine, the Environment and Climate Change Canada stations and muda-clima computed: no fact or number comes from the model. Versão em português.