The rapid and accurate classification of Panax notoginseng leaf diseases is vital for timely disease control and reducing economic losses. Recently, image classification algorithms have shown great promise for plant disease diagnosis, but dataset quantity and quality are crucial. Moreover, classifying P. notoginseng leaf diseases faces severe challenges, including the small features of anthrax and the strong similarity between round spot and melasma diseases. In order to address these problems, we have proposed an ECA-based diffusion model and Inception-SSNet for the classification of the six major P. notoginseng leaf diseases, namely gray mold, powdery mildew, virus infection, anthrax, melasma, and round spot. Specifically, we propose an image generation scheme, in which the lightweight attention mechanism, ECA, is used to capture the dependencies between channels for improving the dataset quantity and quality. To extract disease features more accurately, we developed an Inception-SSNet hybrid model with skip connection, attention feature fusion, and self-calibrated convolutional. These innovative methods enable the model to make better use of local and global information, especially when dealing with diseases with similar features and small targets. The experimental results show that our proposed ECA-based diffusion model FID reaches 42.73, compared with the baseline model, which improved by 74.71%. Further, we tested the classification model using the data set of P. notoginseng leaf disease generation, and the accuracy of 11 mainstream classification models was improved. Our proposed Inception-SSNet classification model achieves an accuracy of 97.04% on the non-generated dataset, which is an improvement of 0.11% compared with the baseline model. On the generated dataset, the accuracy reached 99.44%, which is an improvement of 1.02% compared to the baseline model. This study provides an effective solution for the monitoring of Panax notoginseng diseases.

Effective nitrogen (N) management practices are essential for achieving efficient and sustainable agricultural production. The purpose of this study was to improve N use efficiency (NUE) and minimize N loss by optimizing the rate and type of N fertilizer application while maintaining a high yield of maize. A two-year field experiment with U (urea), S (slow-release N fertilizer), and SU (blending of S and U) under four N application levels (N1: 90 kg ha⁻¹, N2: 120 kg ha⁻¹, N3: 180 kg ha⁻¹, N4: 240 kg ha⁻¹) was conducted to investigate their effects on ammonia (NH₃) volatilization, residual soil nitrate N (${\mathrm{N}\mathrm{O}}_3^{-}-\mathrm{N}$), yield, NUE, apparent N losses of rainfed maize. NH₃ volatilization in SU and S were 38.46% and 16.57% lower than that in U, respectively. SU and S were found to reduce the apparent N losses by 42.98% and 62.23%. SU decreased ${\mathrm{N}\mathrm{O}}_3^{-}-\mathrm{N}$ leaching in deep soils and increased ${\mathrm{N}\mathrm{O}}_3^{-}-\mathrm{N}$ content in topsoil. Compared with U and S, SU significantly increased yield, plant N accumulation, and NUE. SUN4 achieved the maximum maize yield and plant N accumulation, averaging 7968.36 kg ha⁻¹ and 166.45 kg ha⁻¹. In addition, the high yield and NUE were obtained when the mixing ratio of S and U was 53–58% and the N application rate was 150–220 kg ha⁻¹. The findings highlight that SU effectively reduces N losses while ensuring high yield, which could be used as one of the optimal N fertilization strategies for rainfed maize in Northwest China.

The fungal community structure and soil fertility in rhizosphere soil have an important effect on the health of Panax notoginseng (P. notoginseng). The attack of pathogenic fungi and the imbalance of soil fertility can easily lead to diseases. The effect of Bacillus subtilis on improving the community structure of soil fungi has been confirmed, and the corresponding biological agent products have been commercialized. A pot experiment carried out in a greenhouse explored the effect of a biological agent and fertilizer on the fungal community in the rhizosphere of P. notoginseng. In the experiment, fertilization and the addition of biological agents were set up with three gradients, respectively, and the full coupling experiment was adopted, and the blank control group (CK) was set up at the same time. Therefore, there were thirteen treatments in the experiment. NH4 decreased between 36.42% and 11.56%, AP increased between 6.03% and 92.46%, AK increased between 2.99% and 25.40%, TN increased between 0.10% and 9.41%, and TP increased by 18.25% to 47.73% The addition of Bacillus subtilis biological agent decreased the Chao1, Shannon, Simpson, and ACE index of fungi in the rhizosphere soil of P. notoginseng. The Chao1 index decreased between 0.39% and 78.22%; the ACE index decreased between 0.43% and 78.24%. The main pathogenic fungi Cylindrocarpon and Fusarium of P. notoginseng were different in the experimental results. Cylindrocarpon decreased under F1C1, F2C1, and F3C2 treatments, while Fusarium increased under F1C1, F2C2, F3C1, and F3C3 treatments and decreased Fusarium content in rhizosphere soil of P. notoginseng in other treatments. RDA analysis (Redundancy analysis) showed that NH4-N was negatively correlated with the main pathogen Cylindrocarpon, Fusarium, and Ilyonectria, while AP and AK were positively correlated with Cylindrocarpon, Fusarium, and Ilyonectria. The results of the GRA-TOPSIS analysis showed that the score of F3C2 was the highest, while F2C3 and F2C1 ranked second and third, respectively. The calculation results of the theoretical model based on GRA-TOPSIS analysis showed that the GRA-TOPSIS score was highest when the theoretical optimal fertilizer application rate and bacteria application rate were 116.31 kg hm⁻² and 15.83 kg hm⁻², respectively.