Atmospheric turbulence is a key factor contributing to data distortion in mid-to-long-range target observation tasks. Neural networks have become a powerful tool for dealing with such problems due to their strong ability to fit nonlinearities in the spatial domain. However, the degradation in data is not confined solely to the spatial domain but is also present in the frequency domain. In recent years, the academic community has come to recognize the significance of frequency domain information within neural networks. There remains a gap in research on how to combine dual-domain information to reconstruct high-quality images in the field of blind turbulence image restoration. Drawing upon the close association between spatial and frequency domain degradation information, we introduce a novel neural network architecture, termed Dual-Domain Removal Turbulence Network (DDRTNet), designed to improve the quality of reconstructed images. DDRTNet incorporates multiscale spatial and frequency domain attention mechanisms, combined with a dual-domain collaborative learning strategy, effectively integrating global and local information to achieve efficient restoration of atmospheric turbulence-degraded images. Experimental findings demonstrate significant advantages in performance for DDRTNet compared to existing methods, validating its effectiveness in the task of blind turbulence image restoration.

Global change drivers are reshaping agroecosystems and their sustained functions worldwide. While soil microorganisms underpin the resilience of these systems, the individual and interactive effects of multiple anthropogenic stressors on microbial community structure and function using large-scale field experiments remain poorly understood. Here, we utilize a full-factorial field experiment in a subtropical agroecosystem to investigate how land-use intensity, cattle grazing intensity, and altered precipitation regimes interact to shape soil microbiomes. Combining microbiome sequencing with network analyses and functional bioinformatics, we evaluated effects of these drivers on prokaryotic and fungal diversity, composition, predicted functional profiles, and community structure. Land-use intensity emerged as the primary driver of microbial responses, explaining 25% and 13% of the total variation in community composition for prokaryotes and fungi, respectively. Compared to intensively managed pastures, semi-natural pastures had significantly different community composition for prokaryotes and fungi and exhibited 22% higher fungal diversity. Semi-natural pastures were enriched with decomposer-associated taxa and metabolic pathways related to energy and lipid metabolism indicating enhanced microbial activity. Surprisingly, intensively managed pastures showed higher network modularity but lower network richness, suggesting a trade-off between community compartmentalization and complexity under intensive land management. Grazing and precipitation manipulations induced core microbiome changes within land-use intensities but had no impact on overall community structure and no significant interactions with land-use. Together, these findings suggest that long-term land-use legacies exert a persistent influence on soil microbial community structure, function, and organization, shaping the context within which other global change drivers operate in subtropical agroecosystems.