Maps Reveal Where Threatened Himalayan Medicinal Plants Could Survive Climate Change

High in the alpine meadows of Uttarakhand, in the western Himalaya, some of the world’s most valuable medicinal plants are running out of safe ground. Species prized for centuries in traditional medicine are being squeezed by a triple threat: a warming climate, degrading habitats, and relentless harvesting from the wild. Now, a team of researchers has produced something conservationists have long needed — detailed maps that show not only where these plants live today, but where they could realistically be cultivated in the decades ahead. The study, published in Plant Biosystems, combines species distribution modelling with practical feasibility filters to turn raw climate data into actionable conservation planning for seventeen vulnerable medicinal plant species.

The research, led by Neha Thapliyal and K. Chandra Sekar of the G.B. Pant National Institute of Himalayan Environment, together with Rajendra Singh Rawat of the Indian Institute of Geomagnetism, focused on the alpine and subalpine zones of Uttarakhand, a region where plant life is compressed into narrow elevational bands. Mountain species face a particular dilemma under climate change: as temperatures rise, suitable conditions shift upslope, but there is a finite limit to how high a plant can climb. Once the summit is reached, there is nowhere left to go. This phenomenon, sometimes described as the elevator-to-nowhere problem, makes Himalayan flora among the most climate-vulnerable on Earth, and it is precisely why the team chose to model these species at high spatial resolution.

The methodological heart of the study is a stacked species distribution modelling framework, or SSDM. Rather than modelling each species in isolation and simply overlaying the results, the researchers built ensemble models — combining multiple modelling algorithms to reduce the uncertainty inherent in any single approach — and then stacked the outputs for two groups of species. Category-I contained the highly vulnerable species, while Category-II held those classified as vulnerable. Occurrence records were painstakingly compiled from field surveys, herbarium specimens, published literature, and online databases, then cleaned and filtered to yield 471 reliable records. These were modelled against high-resolution bioclimatic variables, the kind of temperature and precipitation surfaces that capture the seasonal rhythms mountain plants depend upon.

The models performed impressively well. The mean area under the curve, a standard measure of model discrimination, reached 0.90 for the Category-I species and 0.88 for Category-II — values that indicate the models distinguish suitable from unsuitable habitat with high reliability. Two climate variables emerged as the dominant drivers for both species groups: precipitation of the coldest quarter and mean temperature of the wettest quarter. In practical terms, this means that winter snowfall and moisture during the growing season, rather than average annual temperature alone, govern where these plants can persist. It is a reminder that for high-altitude flora, the timing of water and warmth matters more than their annual totals, a nuance that coarse global models often miss.

The current-day maps delivered a sobering picture. Only 21.8 percent of climatically suitable habitat for the highly vulnerable species fell within high-suitability classes, and for the vulnerable group the figure dropped to just 13.0 percent. What remains is concentrated between roughly 2,500 and 4,500 metres above sea level — a band of terrain that is both ecologically precious and increasingly pressured by grazing, tourism, and collection. In other words, even before any future climate change is considered, the majority of the climatically appropriate landscape already falls short of the highest suitability thresholds, leaving these species with a narrow and fragmented safety margin.

The future projections sharpened the concern. Under the SSP126 scenario, a relatively optimistic pathway in which emissions decline strongly, suitable habitat for Category-I species declined by 31.3 percent by 2040, with Category-II species losing 25.9 percent. Interestingly, the models showed partial recovery by 2080 under this scenario, suggesting that if the world bends the emissions curve downward, some lost habitat could become climatically suitable again within the century. Under the intermediate SSP245 scenario, however, the story was far grimmer: persistent losses of 67.7 percent for the highly vulnerable group and 64.6 percent for the vulnerable group, with no rebound by 2080. The contrast between these two futures is stark, and it quantifies exactly what is at stake in global climate policy for one small but biologically rich corner of the Himalaya.

What sets this study apart from many climate-impact assessments is its second stage: translating climatic suitability into genuinely cultivable land. A grid cell may have the right temperature and rainfall, but if it sits on a glacier, a steep cliff, a settlement, or beside a busy road, it is useless for establishing cultivation plots. The researchers therefore applied a series of feasibility filters — elevation constraints, land-use and land-cover classification, and proximity to roads as a proxy for human disturbance — and then aggregated the results spatially. The filtering was drastic. Only 23.5 percent of Category-I suitable habitat, roughly 6,495 square kilometres, and 19.2 percent of Category-II habitat, about 6,140 square kilometres, remained feasible for cultivation once the real-world constraints were imposed.

The geography of opportunity that emerged is remarkably concentrated. High- and medium-feasibility zones clustered in four districts: Chamoli, Uttarkashi, Pithoragarh, and Bageshwar. This concentration is good news for planners, because it means that cultivation programmes, nurseries, and community training initiatives can be targeted at a manageable set of landscapes rather than scattered across the entire state. It also aligns with existing conservation infrastructure in the region, including protected areas and biosphere reserves, potentially allowing cultivation zones to complement rather than compete with strict conservation objectives. For the mountain communities who have traditionally harvested these plants from the wild, the maps offer a route to cultivating them instead — reducing pressure on wild populations while securing livelihoods.

The broader significance of the work lies in its demonstration that stacked species distribution models, when coupled with feasibility constraints, can produce guidance that is genuinely actionable rather than merely descriptive. Conservation biology has long been criticised for producing maps that are scientifically elegant but practically inert, identifying habitats that no agency could ever protect or restore. By explicitly asking not just where the climate is right but where cultivation could actually happen, the researchers have bridged the gap between prediction and implementation. The approach is transferable to other mountain regions and other groups of threatened plants, wherever occurrence data and environmental layers are available.

Challenges remain, of course. Species distribution models assume that climate is the primary constraint on distribution, yet biotic interactions, soil chemistry, and dispersal limitations all shape real populations, and the models cannot fully capture these. The occurrence dataset, while carefully curated, reflects sampling patterns that are themselves biased toward accessible terrain. And cultivation at scale brings its own questions about genetic diversity, phytochemical quality, and market dynamics that maps alone cannot answer. Still, as climate change accelerates across the Hindu Kush Himalaya, studies like this one provide exactly the kind of forward-looking, spatially explicit evidence that governments and communities need. For seventeen threatened medicinal plants clinging to the high meadows of Uttarakhand, the maps may represent the difference between a managed transition and a quiet disappearance.

Subject of Research: Habitat suitability and cultivation potential modelling for threatened medicinal plants in the western Himalaya under climate change

Article Title: Modelling and mapping habitat suitability for cultivation potential of threatened medicinal plants of Uttarakhand, west Himalaya, India

Article References: Thapliyal, N., Rawat, R. S., & Sekar, K. C. (2026). Modelling and mapping habitat suitability for cultivation potential of threatened medicinal plants of Uttarakhand, west Himalaya, India. Plant Biosystems, 160(4), Article 230. https://doi.org/10.1007/s44473-026-00220-7

Image Credits: AI Generated

DOI: 10.1007/s44473-026-00220-7

Keywords: medicinal plants, Himalaya, species distribution modelling, climate change, Uttarakhand, habitat suitability, conservation, cultivation, alpine ecosystems, SSDM, biodiversity, plant Biosystems