{"title":"Three Decades of Grassland Emergence in the Terai-Duar Savanna and Grasslands Ecoregion: Patterns and Drivers.","authors":"Swapnil Bhowal, Subham Banerjee, Robert John","doi":"10.1093/aob/mcag051","DOIUrl":null,"url":null,"abstract":"<p><strong>Background and aims: </strong>Grasslands in the Terai-Duar Savanna and Grasslands ecoregion, located at the base of the Himalayas, are disturbance-dependent ecosystems where grassland loss due to woody encroachment has been reported in recent decades. However, some of this loss is offset by grassland emergence, but the extent and drivers of grassland expansion remain poorly understood. Here, we aim to study the relationship between grassland emergence, ecosystem characteristics, and disturbance factors.</p><p><strong>Methods: </strong>We analyzed satellite-based land cover classifications (Landsat TM/OLI) from 1988 to 2019 across eight protected areas (four in each of India and Nepal) to quantify grassland emergence starting from non-grassland land cover. We applied a two-tiered Bayesian modelling framework: (1) a landscape-scale mixed-effects binomial model (with protected area as a random effect) to identify environmental drivers, and (2) site-specific spatial regression models with conditional autoregressive priors to capture local patterns while accounting for spatial autocorrelation.</p><p><strong>Key results: </strong>Over the last three decades, grassland emergence was observed in 9.5% of the non-grassland area, which accounts for ∼37% of the current grassland extent. Grassland emergence was favoured in sites with steep slopes, high soil moisture content (as indicated by the topographic wetness index), and frequent fire, but inhibited at higher elevations and greater distances from human settlements or water bodies. The importance of environmental drivers varied by site, and the effects of fire were strongest in high-rainfall sites, whereas topographic and hydrological factors dominated in the drier sites. Greater grassland formation near human settlements suggests that anthropogenic disturbance (mainly livestock grazing and fires) can mimic natural processes in creating open habitats.</p><p><strong>Conclusions: </strong>These findings underscore the ecological significance of disturbance regimes (flood, human influence, and fire) in maintaining grasslands and habitat management for conservation.</p>","PeriodicalId":8023,"journal":{"name":"Annals of botany","volume":" ","pages":""},"PeriodicalIF":3.6000,"publicationDate":"2026-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Annals of botany","FirstCategoryId":"99","ListUrlMain":"https://doi.org/10.1093/aob/mcag051","RegionNum":2,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"PLANT SCIENCES","Score":null,"Total":0}
引用次数: 0
Abstract
Background and aims: Grasslands in the Terai-Duar Savanna and Grasslands ecoregion, located at the base of the Himalayas, are disturbance-dependent ecosystems where grassland loss due to woody encroachment has been reported in recent decades. However, some of this loss is offset by grassland emergence, but the extent and drivers of grassland expansion remain poorly understood. Here, we aim to study the relationship between grassland emergence, ecosystem characteristics, and disturbance factors.
Methods: We analyzed satellite-based land cover classifications (Landsat TM/OLI) from 1988 to 2019 across eight protected areas (four in each of India and Nepal) to quantify grassland emergence starting from non-grassland land cover. We applied a two-tiered Bayesian modelling framework: (1) a landscape-scale mixed-effects binomial model (with protected area as a random effect) to identify environmental drivers, and (2) site-specific spatial regression models with conditional autoregressive priors to capture local patterns while accounting for spatial autocorrelation.
Key results: Over the last three decades, grassland emergence was observed in 9.5% of the non-grassland area, which accounts for ∼37% of the current grassland extent. Grassland emergence was favoured in sites with steep slopes, high soil moisture content (as indicated by the topographic wetness index), and frequent fire, but inhibited at higher elevations and greater distances from human settlements or water bodies. The importance of environmental drivers varied by site, and the effects of fire were strongest in high-rainfall sites, whereas topographic and hydrological factors dominated in the drier sites. Greater grassland formation near human settlements suggests that anthropogenic disturbance (mainly livestock grazing and fires) can mimic natural processes in creating open habitats.
Conclusions: These findings underscore the ecological significance of disturbance regimes (flood, human influence, and fire) in maintaining grasslands and habitat management for conservation.
期刊介绍:
Annals of Botany is an international plant science journal publishing novel and rigorous research in all areas of plant science. It is published monthly in both electronic and printed forms with at least two extra issues each year that focus on a particular theme in plant biology. The Journal is managed by the Annals of Botany Company, a not-for-profit educational charity established to promote plant science worldwide.
The Journal publishes original research papers, invited and submitted review articles, ''Research in Context'' expanding on original work, ''Botanical Briefings'' as short overviews of important topics, and ''Viewpoints'' giving opinions. All papers in each issue are summarized briefly in Content Snapshots , there are topical news items in the Plant Cuttings section and Book Reviews . A rigorous review process ensures that readers are exposed to genuine and novel advances across a wide spectrum of botanical knowledge. All papers aim to advance knowledge and make a difference to our understanding of plant science.