Reindeer Do Not Migrate According to Weather Apps: Sámi Pastoral Knowledge Corrects What Satellites See in the Arctic
Original Chinese title: 馴鹿不按氣象 App 遷徙：Sámi 放牧知識如何修正衛星看到的北極
Drawing on Sámi reindeer herding practices, snow quality interpretation, Landsat remote sensing, and climate services research, this article examines the complementarity and gaps between satellite data, forecasts, and local knowledge.
Yuan Media AI Editorial Desk
Focuses on science governance, Indigenous knowledge, public technology, and interdisciplinary evidence.
If you know the Arctic from general news, your impression is usually melting ice, starving bears, rising temperatures. These are not wrong, but for Sámi herders who truly depend on reindeer, the problems are far more complex than global average warming. Herding success does not hinge on an abstract warming number; it depends on whether snow is loose or hard, whether a subsurface ice layer has formed, whether lichen can be dug up, if insect pressure appears early, and whether migration paths have been cut by roads, mining, and wind farms. The details of the Arctic often matter more than grand narratives.
Reindeer Need Not Just Forage, but Snow They Can Dig Through to Reach It
Modern agricultural policies often simplify environmental data into temperature, rainfall, snow depth, and similar indicators; for reindeer herders, what is truly fatal is often the structure of the snow layer. Especially in winter, if melting occurs followed by refreezing, a hard crust forms on the surface that makes it difficult for reindeer to dig through the snow to reach lichen. This difference is not easily read from general weather apps, but for herders it can determine seasonal movement and supplementary feeding strategies.
Climate Research on climate change and Sámi reindeer herding has pointed out that climate change is not the only risk; it intertwines with land-use changes, regulations, and infrastructure, shrinking the flexibility herders previously relied upon. As snow conditions become increasingly unstable and extreme events more frequent, conditions become harder to predict from past experience alone; yet without local experience, external models often miss truly useful details.
Satellites Can See Change, But Not Necessarily What It Means
Satellite remote sensing is now standard in Arctic research. It can track snow cover, vegetation, wetlands, road networks, and surface disturbances, making it powerful for understanding large-scale changes. In Remote Sensing of Environment research, researchers combined Landsat imagery with Sámi herders' land-use experience to show how remote sensing depicts landscape change, while local knowledge fills in explanations of migration, seasonal use, and fine-grained environmental states.
This illustrates both the strengths and blind spots of remote sensing. It is excellent at seeing "where things have changed," but not necessarily "how much difference this change makes for whom." A thin road on a map may be a traffic improvement for urban planners, yet for herders it can become an obstacle to reindeer movement, a break point in driving routes, or even alter how snow accumulates due to wind. Looking down from the sky at the Arctic, one easily assumes they see clearly; the real difficulty is understanding how ground life gets rearranged by those changes.
Climate Services That Merely Translate Research Reports Offer Limited Help
In recent years, "climate services" have been frequently mentioned—turning climate data into decision-ready products. For Sámi herding, however, if a climate service only makes weather maps prettier, its practical help may be minimal. Herders often need not seasonal averages but whether freezing rain might occur in the next few days, where snow crust thickens, which migration corridors become muddy due to early melt, or when insect peaks force reindeer to move earlier.
Thus many studies argue climate services should be co-designed with users rather than supplied unilaterally by research institutions. Co-design is not just asking herders to fill out questionnaires; it involves letting them define which environmental signals truly matter, at what spatial scale they should be presented, where risk thresholds lie, and when scientific forecasts should be weighed alongside oral traditions. Without this step, climate services resemble many official information products: professionally looking yet far removed from the field.
"Green Transition" May Also Make Herding More Vulnerable
When discussing Arctic risks, people often treat fossil fuels and warming as the primary enemies, rarely addressing that the green transition itself can bring pressure. Wind farms, mining development, roads, and power transmission lines frequently enter northern landscapes under the banner of sustainability; yet for reindeer herding, spatial fragmentation sometimes poses a more direct threat than average temperature rise. Wind farms are not just visual changes—they may introduce noise, infrastructure roads, human traffic, and avoidance effects. Mining is not merely surface extraction; it can also cause migration route detours and cumulative stress.
What deserves most vigilance here is that modern policy often assumes it is solving the global optimum: decarbonization is good, green energy is good, so local communities should cooperate. The problem is if the costs fall disproportionately on herders already highly dependent on land flexibility, then this sustainability narrative exhibits selective blindness. It sees the Earth but not the place.
Herder Knowledge Is Not "Case Material" But a Diagnostic System
Sámi knowledge is often treated by external researchers as supplementary case material: certain vocabulary terms are interesting, some stories inspiring, and experiences useful for model validation. For herders, however, these are not scattered bits of knowledge but an entire diagnostic system. It includes snow classification, animal behavior, past routes, seasonal expectations, insect pressure, predator risk, and community coordination. Breaking it into cute fragments only leads people to mistake local knowledge as colorful footnotes.
True Two-Eyed Seeing collaboration should recognize this as a system already possessing logic, precision, and practical consequences. Science can dialogue with it but should not assume its own superiority. Otherwise, even advanced remote sensing merely turns local experience into calibration parameters, then attaches the results to its own name.
What Taiwan Can Learn from This
The Sámi context differs from Taiwan's Indigenous pastoralism, hunting, or highland agriculture; direct application is impossible. Yet methodologically it offers strong inspiration: first, environmental services that do not start from users' real risks often produce information that looks beautiful but decision-making goes off-focus. Second, satellites and local knowledge are not alternatives but different-scale observation systems. Third, policy cannot focus only on grand goals; it must also consider how spatial fragmentation alters on-the-ground action flexibility.
Today many technology governance initiatives boast abundant data, strong models, accurate forecasts. Sámi herding knowledge reminds us that truly usable knowledge is not necessarily the most data-rich but rather the best at responding to life decisions. Reindeer do not migrate according to weather apps; humans should not rely solely on abstract models to understand the north. Satellites see the Arctic from above; herders live within it; if they cannot speak together, only a governance illusion that looks modern yet fails in practice remains.
Supplementary Observation: Climate Services Must Address Decisions, Not Just Weather
For reindeer pastoralism, the same average temperature can correspond to completely different feasible routes. After snowfall, brief melting followed by refreezing may create hard snow crusts animals cannot dig through; roads, mining areas, or wind farms may cut off alternative paths that were previously usable. If climate services only provide regional averages, they miss the questions herders truly need answered: which road segment is passable when, where lichen remains accessible, and what choices remain during sudden changes.
Therefore service design should split time scales. Satellites suit large-scale surface comparison and long-term change; ground observation can supplement snow layers, ice crusts, wind erosion, and animal behavior in real time; herding records explain how these variations affect actual decisions. The three are not substitutes but must be marked within the same workflow with data dates, resolutions, errors, and final confirmers. When information conflicts, channels should remain for on-site users to explain differences.
Data governance is also part of climate services. Migration routes, seasonal stopover points, and herding strategies may involve land rights, commercial development pressures, and community safety; technical ability to locate does not mean default openness. Projects must first agree who can see raw data, how much the public version should blur, how research ends with preservation or deletion, and who corrects error markings. Without these arrangements, so-called co-production risks turning local knowledge into new external data assets.
In public communication, Sámi experience should not be written as a single Arctic story. Different countries, grazing areas, families, and governance systems face distinct pressures; the article offers understanding methods, not directly applicable policy recipes. For Taiwan readers, the key inspiration is that any environmental service must embed local users' decision questions, data rights, and feedback mechanisms into its design rather than inviting community trial only after platform completion.
Sources retained from the Chinese original
AI use and content-safety disclosure
This article was assisted by AI for data organization, structural drafting, and sentence polishing; human editors set viewpoints and fact-checking directions