Monitoring tropical and subtropical forest soil CO2 emission efflux (FSCO2) is crucial for understanding the global carbon cycle and terrestrial ecosystem respiration. In this study, we addressed the challenge of low spatiotemporal resolution in FSCO2 monitoring by combining data fusion and model methods to improve the accuracy of quantitative inversion. We used time series Landsat 8 LST and MODIS LST fusion images and a linear mixed effect model to estimate FSCO2 at watershed scale. Our results show that modeling without random factors, and the use of Fusion LST as the fixed predictor, resulted in 47% (marginal R2 = 0.47) of FSCO2 variability in the Monthly random effect model, while it only accounted for 19% of FSCO2 variability in the Daily random effect model and 7% in the Seasonally random effect model. However, the inclusion of random effects in the model’s parameterization improved the performance of both models. The Monthly random effect model that performed optimally had an explanation rate of 55.3% (conditional R2 = 0.55 and t value > 1.9) for FSCO2 variability and yielded the smallest deviation from observed FSCO2. Our study highlights the importance of incorporating random effects and using Fusion LST as a fixed predictor to improve the accuracy of FSCO2 monitoring in tropical and subtropical forests.
Quantifying dust aerosol radiative impacts on crop growth in arid regions is challenging due to sparse ground observation networks for photosynthetically active radiation (PAR). Conventional meteorological stations only provide regional-averaged solar radiation and fail to capture fine spatial heterogeneity and instantaneous attenuation caused by dust storms. To address this gap, this study developed a coupled framework integrating WRF-Chem, LibRadtran, multi-source remote sensing, and interpretable machine learning. We combined field sampling data, remote sensing products, and atmospheric simulations to explore how dust aerosol direct radiative forcing alters cotton gross primary productivity (GPP) and yield across the Weigan River Basin, Xinjiang, China. Results revealed significant PAR reduction induced by dust in 87% of cotton fields (p < 0.05). Dust presented a dual effect: it reduced photosynthetic productivity via radiation attenuation, while alleviating heat stress above 35 °C. SHAP analysis demonstrated that cotton GPP and yield declined nonlinearly when daily PAR loss exceeded 20 W·m−2, with the flowering-to-boll stage (July–August) identified as the most sensitive phenological window. This study verifies the necessity of combining atmospheric models and remote sensing for fine-scale assessment of dust radiative effects in data-scarce regions. The identified threshold provides practical references for targeted field management in arid cotton areas.