Feng et al. (2026) Climate-Based Natural Suitability Index (CNSI) for Blueberry Cultivation in China: Spatiotemporal Evolution and Influencing Factors

Identification

• Journal: Agronomy
• Year: 2026
• Date: 2026-01-15
• Authors: Yixuan Feng, Hu Chen, Jiayi Liu, Xinchun Wang, Jinying Li, Ying Wang, Junnan Wu, Lin Wu, Yanan Li
• DOI: 10.3390/agronomy16020211

Research Groups

• College of Horticulture, Jilin Agricultural University, Changchun, China
• Jilin Academy of Agricultural Sciences, Changchun, China

Short Summary

This study developed a Climate-Based Natural Suitability Index (CNSI) for blueberry cultivation in 19 Chinese provinces (2008-2023) to analyze its spatiotemporal evolution and driving factors, finding that CNSI shows moderate convergence and is primarily influenced by irrigation and fertilizer inputs, with significant neighborhood effects.

Objective

• To characterize the spatiotemporal evolution and identify the driving mechanisms of blueberry climatic suitability realization in 19 major producing provinces in China.

Study Configuration

• Spatial Scale: 19 major blueberry-producing provinces in China, divided into three subregions: Northern Dryland Belt, Middle and Lower Yangtze Rice Belt, and Southern Subtropical and Southwestern Mountainous Belt.
• Temporal Scale: 2008–2023

Methodology and Data

• Models used:
  • Climate-Based Natural Suitability Index (CNSI) framework (ratio of observed production shares to climate-based benchmark shares).
  • Climate-only Fixed-Effects Poisson Pseudo-Maximum Likelihood (FE–PPML) model.
  • Three-dimensional Kernel Density Estimation (KDE).
  • Traditional Markov Chains.
  • Spatial Markov Chains (using Queen contiguity).
  • Optimal Parameter GeoDetector (OPGD) for factor and interaction detection.
• Data sources:
  • Climatic indicators: Chilling portions (CP), growing degree days above 7 °C (GDD7), and water balance (WB) derived from ERA5-Land reanalysis data and TerraClimate reanalysis data.
  • Socioeconomic and industrial support factors: Effective facility irrigation area, fertilizer application rate, agricultural labor force, non-agricultural employment opportunities, urbanization rate, per-capita fruit consumption, per-capita express delivery volume, and comprehensive transportation accessibility from provincial statistical yearbooks, China Rural Statistical Yearbook, industry statistical bulletins, postal and transportation yearbooks, and National Bureau of Statistics.
  • Blueberry production: Provincial annual blueberry production data.

Main Results

• The national CNSI exhibited a stable, moderately right-skewed unimodal distribution from 2008 to 2023, with partial convergence and a narrowing interprovincial gap.
• Suitability realization was highest in the middle and lower Yangtze River rice-growing belt, while the northern dryland belt and the southern subtropical–southwestern mountainous belt showed persistent mismatches between climatic potential and production advantages.
• Markov chain analysis revealed strong path dependence in CNSI states (retention probabilities for S1 and S4 around or above 0.7) and moderate overall mobility (Shorrocks index of 0.412, Bartholomew index of 1.39).
• Spatial Markov analysis indicated coexisting "low–low lock-in" and "high–high club" phenomena reinforced by neighborhood effects, with the highest state mobility observed in medium-neighborhood scenarios.
• GeoDetector results identified effective facility irrigation area (q-values up to 0.65), fertilizer input (q-values up to 0.70), and agricultural labor (q-values up to 0.62) as the dominant factors explaining spatial variations in CNSI.
• Comprehensive transportation accessibility (q-values up to 0.27), urbanization rate (q-values up to 0.14), and per-capita fruit consumption (q-values up to 0.33) served as important complementary factors.
• Multi-factor interactions, particularly irrigation-centered combinations (e.g., X1 × X5, X1 × X6, X1 × X7, X1 × X8), demonstrated strong dual-factor enhancement or near-nonlinear enhancement, with interaction q-values often exceeding 0.60 in later periods.

Contributions

• Developed a novel Climate-Based Natural Suitability Index (CNSI) using a climate-only fixed-effects Poisson Pseudo-Maximum Likelihood (PPML) model to quantify the realization of climatic potential relative to observed production shares.
• Revealed the long-term spatiotemporal evolutionary trajectory of blueberry climatic suitability in China using a comprehensive province–year panel (2008–2023), providing empirical evidence for understanding spatial restructuring under climate constraints.
• Applied the Optimal Parameter GeoDetector (OPGD) to identify key socioeconomic factors and quantify their interactions driving the conversion of climatic potential into competitive production shares, offering a basis for differentiated adaptive management strategies.

Funding

• Jilin Provincial Natural Science Foundation (20250102302JC)
• Jilin Provincial Fruit and Vegetable Industry Technology System (JLARS-2025-120101)
• Jilin Rural Revitalization Think Tank Project: Research on the Development Path and Countermeasures of Characteristic Fruit and Vegetable Industry in Jilin Province
• Research on the “Jingzi Model” of University-Local Cooperation under the Background of Rural Revitalization

Citation

@article{Feng2026ClimateBased,
  author = {Feng, Yixuan and Chen, Hu and Liu, Jiayi and Wang, Xinchun and Li, Jinying and Wang, Ying and Wu, Junnan and Wu, Lin and Li, Yanan},
  title = {Climate-Based Natural Suitability Index (CNSI) for Blueberry Cultivation in China: Spatiotemporal Evolution and Influencing Factors},
  journal = {Agronomy},
  year = {2026},
  doi = {10.3390/agronomy16020211},
  url = {https://doi.org/10.3390/agronomy16020211}
}