%0 Dataset
%T Sample data set of moisture survey in the active layer of the permafrost region of the Qinghai-Tibet Plateau (2009-2024)
%J National Cryosphere Desert Data Center
%I National Cryosphere Desert Data Center(www.ncdc.ac.cn)
%U http://www.ncdc.ac.cn/portal/metadata/201d1d8b-47bb-414a-9120-baf90be91e1c
%W NCDC
%R 10.12072/ncdc.permafrost.db7703.2026
%A Du erji
%K Qinghai-Tibet Plateau;permafrost region;active layer moisture
%X This dataset was systematically collected and compiled for the permafrost active layer on the Qinghai–Tibet Plateau during 2009–2024, supported by the Fundamental Work Special Program of the Ministry of Science and Technology of China (2008FY110200), the Second Qinghai–Tibet Plateau Scientific Expedition and Research Program (2019QZKK020108), the National Natural Science Foundation of China (41301066, 41931180, 42525105), and the Strategic Priority Research Program of the Chinese Academy of Sciences (2025NCDC010). The dataset records the average volumetric water content (VWC) within the active layer depth at each survey site. All field surveys were conducted annually from September to October, corresponding to the maximum thawing period of seasonally frozen soil, to ensure spatiotemporal consistency of observations. A total of 342 valid survey sites are included in the dataset. Among them, 123 sites adopted the field test pit excavation and ring-knife sampling method with a 10 cm vertical interval to obtain in-situ measured active-layer VWC. The remaining 219 sites adopted ground-penetrating radar (GPR) detection. The soil dielectric constant (ε) was first retrieved from GPR signals, and the active-layer VWC was subsequently calculated using the empirical modelθ=0.458^0.26-0.664. To quantitatively evaluate the reliability and intrinsic uncertainty of GPR-retrieved VWC, paired validation was performed based on 18 collocated calibration pits with VWC ranging from 0.124 to 0.614 m³·m⁻³. The validation results yielded a mean absolute error (MAE) of 0.027, a root mean square error (RMSE) of 0.032, and a negligible bias of +0.0003. The Pearson correlation coefficient, coefficient of determination (R²), and Lin’s concordance correlation coefficient were 0.974, 0.9484, and 0.9726, respectively. The 95% Bland–Altman limits of agreement were [−0.0646, +0.0652] m³·m⁻³. The paired t-test (t = 0.0356, p = 0.9708) confirmed no statistically significant systematic difference betwe