Ma et al. (2026) Infrared land surface emissivity dynamics in the Taklimakan desert from 2001 to 2023

Identification

• Journal: Scientific Reports
• Year: 2026
• Date: 2026-01-15
• Authors: Yufen Ma, Ali Mamtimin, Kang Zeng, Ailiyaer Aihaiti, Junjian Liu, Zonghui Liu
• DOI: 10.1038/s41598-025-31933-y

Research Groups

• Institute of Desert Meteorology, China Meteorological Administration, Urumqi, China
• National Observation and Research Station of Desert Meteorology, Taklimakan Desert of Xinjiang, Urumqi, China
• Taklimakan Desert Meteorology Field Experiment Station of China Meteorological Administration, Urumqi, China
• Xinjiang Key Laboratory of Desert Meteorology and Sandstorm, Urumqi, China
• Tazhong National Climate Observatory, Qiemo, China

Short Summary

This study quantifies the spatiotemporal dynamics of infrared land surface emissivity (LSE) in the hyper-arid Taklimakan Desert from 2001 to 2023, revealing a paradoxical LSE increase despite regional drying and warming, primarily driven by thermal-aeolian processes with distinct wavelength-dependent responses.

Objective

• To quantify the long-term (2001–2023) spatiotemporal dynamics of LSE at 8.3 μm, 8.6 μm, and 9.1 μm across the Taklimakan Desert.
• To disentangle the independent and synergistic controls of soil moisture decline and surface temperature rise on LSE using a hybrid statistical and machine learning framework.
• To provide wavelength-specific calibration coefficients to refine land surface temperature retrievals for future satellite missions (e.g., ESA LSTM).
• To test the hypothesis that thermal-aeolian processes dominate LSE variability in the Taklimakan Desert once soil moisture falls below a critical threshold (approximately 2 g kg⁻¹), superseding hydrological controls.
• To test the hypothesis that the magnitude and sign of LSE response to temperature are wavelength-dependent, with stronger negative responses at longer wavelengths (e.g., 9.1 μm).

Study Configuration

• Spatial Scale: Taklimakan Desert, analyzed at a 5 km grid cell resolution.
• Temporal Scale: 23-year period, from 2001 to 2023, with monthly and interannual analyses.

Methodology and Data

• Models used: Mann-Kendall trend analysis, Sen’s slope estimator, partial correlation analysis, Random Forest (RF, 1000 trees, 10-fold cross-validation), Multiple Linear Regression (MLR), Theil-Sen estimators, one-way Analysis of Variance (ANOVA), Levene’s test, Shapiro-Wilk test, Tukey’s Honest Significant Difference (HSD) tests, Getis-Ord’s statistics, Variance Inflation Factor (VIF) checks.
• Data sources:
  • Satellite observations: CAMEL Earth System Data Record (ESDR) V003 (2001–2023) for global monthly LSE at 5 km spatial resolution (8.3 μm, 8.6 μm, 9.1 μm), integrating UW-Madison MODIS and JPL ASTER GEDv4.
  • Reanalysis data: ERA5 (ECMWF) for monthly averaged Skin Reservoir Content (SRC, soil moisture in top 7 cm layer) and Skin Temperature (SKT) at 0.25° spatial resolution.
  • Ancillary data: MODIS MCD12Q1 annual product (IGBP scheme) for land cover (500 m resolution), Shuttle Radar Topography Mission (SRTM) digital elevation model (90 m resolution, aggregated to 5 km), and Harmonized World Soil Database (HWSD) v2.0 for soil texture (clay fraction).

Main Results

• Land surface emissivity (LSE) paradoxically increased by 0.053 to 0.062 per decade (2001–2023) despite regional drying (-0.15 kg m⁻² decade⁻¹) and warming (+0.31 °C decade⁻¹). Thermal-aeolian processes explained 68 ± 7% of this variance.
• Surface temperature independently reduced LSE by an average of 0.0029 ± 0.0012 per °C, with the maximum effect observed at the 9.1 μm band (-0.0035 ± 0.0015 per °C).
• LSE responses exhibited distinct wavelength-dependent characteristics: the 9.1 μm band showed the highest interannual stability (no significant trend, p > 0.05), while the 8.3 μm band displayed the greatest spatial heterogeneity (ΔLSE > 0.07 in central dunes).
• The positive correlation between soil moisture (SRC) and LSE weakened over time, with correlation coefficients declining by approximately 0.15 from 2001 to 2023, and standardized MLR coefficients for SRC decreasing by 35–40% after 2010.
• The relative importance of surface temperature (SKT) as a driver increased post-2010, rising from 25–35% to 35–45% in Random Forest models, indicating a shift towards thermal dominance.
• Interaction analysis revealed antagonistic, non-linear relationships between SRC and SKT, with consistently negative interaction terms in Multiple Linear Regression models.
• Sensitivity scenarios demonstrated that LSE changes due to SKT variation were 2.3–3.1 times greater than those simulated under SRC variation, reinforcing the growing dominance of thermal-aeolian processes.

Contributions

• Provides the first long-term (23-year), multi-decadal analysis of LSE dynamics in a hyper-arid desert, extending beyond previous short-term studies.
• Conducts a systematic, wavelength-specific (8.3 μm, 8.6 μm, 9.1 μm) investigation, demonstrating that driver responses are critically dependent on the spectral band.
• Identifies and quantifies a paradoxical trend of increasing LSE despite regional drying and warming, attributing it to a previously underappreciated thermal-aeolian coupling mechanism, thereby challenging existing soil moisture-centric LSE paradigms.
• Offers critical constraints for desert–climate feedback models and provides actionable, wavelength-specific calibration coefficients (-0.0035 LSE °C⁻¹ at 9.1 μm) to improve future space-borne land surface temperature retrievals (e.g., ESA LSTM).
• Reconciles contradictions between previous short-term field observations and laboratory studies regarding LSE sensitivity to temperature and mineralogy.

Funding

• The Joint Research Project for Meteorological Capacity Improvement of CMA (Grant No. 24NLTSZ006)
• Tianshan Talent Training Program - Youth Talent Support Project (Grant No. 2024TSYCQNTJ0002)
• Youth Innovation Team of China Meteorological Administration (Grant No. CMA2024QN13)
• “Tianshan Talent” Training Program–Science and Technology Innovation Team (Tianshan Innovation Team) Project (Grant No. 2022TSYCTD0007)
• Innovation and development project of Xinjiang Meteorological Service (Grant No. ZD202306)
• S&T Development Fund of CAMS (Grant No. 2021KJ034)
• Xinjiang Key Laboratory of Desert Meteorology and Sandstorms Award Funding (Grant No. 2023-38)

Citation

@article{Ma2026Infrared,
  author = {Ma, Yufen and Mamtimin, Ali and Zeng, Kang and Aihaiti, Ailiyaer and Liu, Junjian and Liu, Zonghui},
  title = {Infrared land surface emissivity dynamics in the Taklimakan desert from 2001 to 2023},
  journal = {Scientific Reports},
  year = {2026},
  doi = {10.1038/s41598-025-31933-y},
  url = {https://doi.org/10.1038/s41598-025-31933-y}
}