Zhao et al. (2025) A new pattern expanding current temperature models: A negative correlation between soil respiration and temperature in cold environments

Identification

• Journal: CATENA
• Year: 2025
• Date: 2025-11-18
• Authors: Wei Zhao, Meng Yang, Qiufeng Wang, Tianxiang Hao, Jianxing Zhu, Zhi Chen, Guirui Yu
• DOI: 10.1016/j.catena.2025.109653

Research Groups

• Key Laboratory of Ecosystem Network Observation and Modeling, Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing 100101, China
• College of Resources and Environment, University of Chinese Academy of Sciences, Beijing 100049, China

Short Summary

This study investigates the relationship between soil respiration (Rs) and temperature across China, revealing a novel U-shaped annual Rs-temperature relationship with a negative correlation in cold regions (mean annual temperature < 3.75 °C) due to moisture-mediated suppression. This finding highlights that traditional models significantly underestimate Rs in these cold environments and underscores the importance of scale-dependent modeling for accurate carbon flux predictions.

Objective

• To analyze the relationship between soil respiration (Rs) and temperature, specifically examining whether the relationship in low-temperature regions exhibits a positive correlation.
• To investigate the mechanisms underlying the observed negative correlations between Rs and temperature in cold environments.
• To evaluate the effect of this negative correlation on regional Rs assessments and predictions.

Study Configuration

• Spatial Scale: Terrestrial ecosystems across China, utilizing 2279 annual and 9723 monthly in situ Rs observation records. Environmental factor data were at spatial resolutions of 0.1° (temperature), 0.05° (soil moisture), 0.08° (Leaf Area Index), and 0.05° (soil organic carbon).
• Temporal Scale: Annual and monthly Rs observations. Environmental data covered 1979–2018 for temperature, 2002–2018 for soil moisture, and 8-day resolution for Leaf Area Index (aggregated to annual).

Methodology and Data

• Models used:
  • Standard exponential (Van’t Hoff)
  • Composite exponential-quadratic (U-shaped/Gaussian)
  • Arctangent (Sigmoid)
  • Rate decay exponential (Lloyd–Taylor)
  • Locally estimated scatterplot smoothing (LOESS)
  • Structural Equation Modeling (SEM) using the "piecewiseSEM" package.
• Data sources:
  • Publicly available in situ soil respiration (Rs) observation data in Chinese terrestrial ecosystems (compiled up to 2022).
  • China Meteorological Forcing Dataset (1979–2018) for temperature.
  • Soil moisture content dataset (2002–2018) from http://poles.tpdc.ac.cn/.
  • GLOBMAP LAI Version 3 from https://www.resdc.cn/ for Leaf Area Index (LAI).
  • Regridded Harmonized World Soil Database V1.2 from https://gaez.fao.org/pages/hwsd for soil organic carbon (SOC).
  • Chinese data from the monthly global Rs dataset (Jian et al., 2018) for monthly Rs.

Main Results

• Annual Rs exhibited a U-shaped relationship with mean annual temperature (MAT) across Chinese terrestrial ecosystems, with an inflection point at 3.75 °C.
• In cold regions (MAT < 3.75 °C), primarily the Qinghai–Tibet Plateau and Northeast Cold Temperate Zone, annual Rs was negatively correlated with temperature.
• This unexpected inverse pattern was primarily attributed to the inhibitory effect of increased soil moisture associated with rising MAT, which suppresses microbial activity.
• Traditional models (Van’t Hoff, Lloyd–Taylor, and Arctangent) systematically underestimated annual Rs in these cold regions by up to 32 %.
• In contrast, monthly Rs consistently increased with temperature and showed no negative correlation, indicating a scale-dependent divergence.
• The U-shaped model predicted mean Rs values for China that were 1 %, 4 %, and 9 % higher than those of the Arctangent, Lloyd–Taylor, and Van’t Hoff models, respectively.
• In cold regions (MAT < 3.75 °C), the U-shaped model predicted mean Rs values that were 5 %, 15 %, and 32 % higher than those of the Arctangent, Lloyd–Taylor, and Van’t Hoff models, respectively.
• Scale-dependent differences between annual and monthly Rs–temperature responses were explained by Jensen’s inequality and spatial heterogeneity in Rs responses along climatic gradients.

Contributions

• Provides the first regional-scale evidence of a significant negative annual Rs–temperature correlation in cold environments, challenging the conventional view of a monotonic positive relationship.
• Clarifies the mechanistic basis of this negative correlation, attributing it to moisture-mediated suppression of microbial activity.
• Demonstrates that traditional Rs–temperature models systematically underestimate soil carbon efflux in cold regions, leading to biases in regional and global carbon budget assessments.
• Explains the scale-dependent divergence between annual and monthly Rs–temperature responses by highlighting the roles of Jensen’s inequality and spatial heterogeneity.
• Offers insights into why annual-scale Rs estimates have been consistently reported as higher than monthly-scale estimates.
• Underscores the critical need for refined multiscale models and non-linear transformations for scale conversion to improve the accuracy of soil carbon flux predictions under climate change.

Funding

• National Natural Science Foundation of China (Grant No. 42261144688)
• National Key R&D Program of China (Grant No. 2022YFF1301801)
• National Natural Science Foundation of China (Grant No. 31988102)
• National Natural Science Foundation of China (Grant No. 31800406)

Citation

@article{Zhao2025new,
  author = {Zhao, Wei and Yang, Meng and Wang, Qiufeng and Hao, Tianxiang and Zhu, Jianxing and Chen, Zhi and Yu, Guirui},
  title = {A new pattern expanding current temperature models: A negative correlation between soil respiration and temperature in cold environments},
  journal = {CATENA},
  year = {2025},
  doi = {10.1016/j.catena.2025.109653},
  url = {https://doi.org/10.1016/j.catena.2025.109653}
}