Century-long analysis uncovers systemic sampling biases in subtropical mountain herbaria, guiding targeted biodiversity conservation
South China Botanical Garden, Chinese Academy of Sciences
image: Herbarium collections from subtropical mountains exhibit indicate regional, temporal, taxonomic, and collector biases.
Credit: Hanxi Wu, Kuiling Zu, Fusheng Chen, Yuanqiu Liu, Wenyuan Zhang, Jianjun Li, Wenping Deng, Lingjian Tao, and Xin Ning
Date: June 18, 2026
Nanchang, China: Herbarium specimens serve as the empirical foundation of biodiversity research, underpinning species distribution mapping, ecological restoration, and conservation policy-making. However, global herbarium records are well documented to carry inherent geographic, temporal, and taxonomic biases that can distort macroecological conclusions and mislead conservation prioritization. Subtropical mountain ecosystems, as global hotspots of vascular plant diversity, remain understudied regarding these systematic biases, leaving critical gaps in regional flora knowledge.
To fill this gap, a research team from Jiangxi Agricultural University conducted a comprehensive analysis of plant specimen records spanning 1900 to 2019 across nine representative subtropical mountains in Jiangxi Province. Using data from national specimen infrastructure and virtual herbarium platforms, the team standardized taxonomic names, georeferenced collection sites, and integrated mountain attributes including latitude, elevation, area, and research intensity. The Chao1 richness estimator, multiple linear regression, and correlation analysis were applied to quantify biases across four dimensions: geography, time, taxonomy, and collector contribution.
The findings reveal pronounced geographic imbalance. Northwestern mountains such as Mt. Lushan, Mt. Jinggang, and Mt. Wugong have substantially higher collection density, while southern sites including Mt. Jiulian and Mt. Qiyun are chronically undersampled. Statistical modeling confirms that latitude, longitude, research intensity, and relative elevation are all significant drivers of collection volume, shaped by transportation accessibility, historical research legacy, and topographic complexity.
Temporally, collection activity features two distinct historical peaks in 1958–1959 and 1963–1965, coinciding with national botanical surveys and regional flora compilation projects. A third phase of sustained growth began in the early 2000s, fueled by strengthened national ecological protection policies and expanded research investment. Seasonally, collections concentrate heavily between April and November, creating phenological bias that underrepresents winter plant traits and dormancy stages.
Taxonomically, angiosperms dominate specimen holdings, and endangered, endemic, and alien species receive disproportionately greater sampling attention, leading to uneven representation across plant groups. Most notably, collector concentration is extreme: the top 10% of contributors account for more than half of all specimens across the nine mountains, a “botanist effect” that amplifies all other sampling biases.
The authors note that these interconnected biases undermine the reliability of herbarium data for climate change impact assessments and conservation planning. To address these gaps, they recommend four targeted measures: prioritizing field surveys in undersampled southern mountains, implementing year-round sampling to reduce seasonal bias, expanding collection coverage of underrepresented taxa, and engaging citizen scientists and local communities to diversify collector bases.
“This century-scale diagnosis enables us to move beyond unrecognized data limitations and build more robust, inclusive biodiversity monitoring systems,” the authors conclude. “The findings offer a scientific blueprint for optimizing subtropical mountain floristic surveys and advancing evidence-based conservation in a core ecological functional zone of China.”
Original Source
Wu, Hanxi, Kuiling Zu, Fusheng Chen, et al. 2025. “Herbarium Collection Biases in Subtropical Mountains in Jiangxi, China Over the Past Century,” Biological Diversity: 2(4): 206–217.
https://onlinelibrary.wiley.com/doi/10.1002/bod2.70017
About the Author
Hanxi Wu (First author), undergraduate student majoring in Forestry, Jiangxi Agricultural University.
Kuiling Zu (Corresponding author), Associate Professor, College of Forestry / College of Landscape Architecture and Art, Jiangxi Agricultural University. Her core research focus is biodiversity conservation. She has served as Principal Investigator for 7 related scientific research projects, and her research findings have been published in leading journals including Global Change Biology, New Phytologist, and Global Ecology and Biogeography.
Fusheng Chen (Corresponding author), Professor & Doctoral Supervisor, College of Forestry / College of Landscape Architecture and Art, Jiangxi Agricultural University. Professor Chen specializes in silviculture, forest ecology, and soil and water conservation, with focused research on forest soils, root and leaf traits, planted forest cultivation, and natural forest conservation. He has published 124 peer-reviewed academic papers, which have garnered 2,864 total citations, with an H-index of 28.
About the Journal
Biological Diversity (ISSN: 2994-4139) is a peer-reviewed, international, open-access journal sponsored by the South China Botanical Garden (SCBG), Chinese Academy of Sciences (CAS), and published in partnership with John Wiley & Sons Australia, Ltd. It is dedicated to advancing biodiversity conservation, safeguarding ecosystem functions and services, and promoting the sustainable utilization of biological resources under global environmental change. The journal welcomes original research, reviews, commentaries, and short communications across a broad spectrum of disciplines, including botany, zoology, microbiology, taxonomy, phylogenetics, genomics, cytology, ecology, climatology, economics, sociology, and real-time policy theory.
Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.