Threshold-driven soil degradation and productivity decline under long-term tea monoculture in tropical highlands: Evidence from field data and predictive modelling

This dataset provides a comprehensive collection of soil chemical, physical, and biological properties measured under long-term tea (Camellia sinensis) monoculture systems in the tropical highlands of Vietnam. The dataset was developed to investigate soil degradation dynamics and identify critical threshold values associated with productivity decline. Soil samples were collected from four land-use types, including native forest (reference system) and tea plantations of different ages (5, 10, and 20 years). Sampling was conducted at a depth of 0–20 cm with five replicates per site to ensure statistical robustness. The study area is located in Bao Loc, Lam Dong Province (11.54°N, 107.45°E), characterized by Ferralsols derived from basalt under a tropical monsoon climate. Measured variables include key indicators of soil quality: soil organic carbon (SOC), available phosphorus (P), available potassium (K), bulk density, plant-available water capacity (PAWC), soil pH, and earthworm density. In addition, tea yield data were included to quantify soil–productivity relationships. The dataset supports multiple levels of analysis, including descriptive statistics, regression modelling, and threshold (piecewise) analysis. It enables the identification of critical thresholds, particularly SOC (~12 mg g⁻¹) and available P (~6 µg g⁻¹), below which soil functionality and crop productivity decline sharply. These findings highlight the non-linear, threshold-driven nature of soil degradation in perennial monoculture systems. This dataset is designed to be fully reproducible and reusable. It includes raw data, processed datasets for regression modelling, and threshold analysis subsets. Accompanying files provide detailed metadata, documentation, and analytical scripts (R), allowing users to replicate the results and extend the analysis. The dataset is relevant for researchers working in soil science, agroecosystem sustainability, land degradation, and climate change adaptation. It can also support modelling studies, meta-analyses, and the development of soil health indicators for tropical agricultural systems. License: Creative Commons Attribution 4.0 International (CC BY 4.0).