Leaf area index (LAI) is a critical indicator of canopy architecture and physiological performance, serving as a key parameter for crop growth monitoring and management. Although UAV multispectral imagery provides rich spectral and spatial information, the limitations of single texture features for LAI estimation still require further exploration. To address this issue, this study developed a multi-source feature fusion framework that integrates vegetation indices (VIs), texture features (TFs), and texture indices (TIs) within a stacked ensemble approach combining Partial Least Squares Regression (PLSR) with Support Vector Machine (SVM), Random Forest (RF), and Gradient Boosting Decision Tree (GBDT) algorithms to estimate maize LAI.A field experiment was conducted under three planting densities (42,000, 63,000, and 84,000 plants ha⁻¹) and four nitrogen rates (0, 80, 160, 240 kg N ha⁻¹) to assess the potential of UAV-based multispectral imagery for maize LAI estimation. The results show that when using partial least squares regression (PLSR) combined with RF, SVM and GBDT to estimate maize LAI, the R² values are 0.653, 0.697 and 0.634, and the RMSE is 0.650, 0.608 and 0.668, respectively, when only vegetation indices (VIs) is used as input. After texture features (TFs) incorporation, the R² increases to 0.717, 0.794, and 0.801, and the RMSE decreases to 0.587, 0.500, and 0.492. Further inclusion of the texture indices (TIs) raises the R² to 0.789, 0.804, and 0.844, with RMSE of 0.506, 0.489, and 0.436, respectively. Independent test set validation under contrasting conditions confirmed that our multi-model fusion framework (PLSR+GBDT) with multi-source feature fusion (VIs+TFs+TIs) effectively estimated LAI, achieving an R² of 0.859 and 0.794. These results demonstrate that multi-source feature integration via machine learning enables robust and accurate estimation of maize LAI, providing a valuable tool for precision agriculture and crop growth monitoring.

To explore the response of soil organic carbon (SOC) and carbon components to surface coverage in wheat fields in semi-arid rainfed regions, a two-year field experiment was conducted under three treatments: straw strip mulching (SM), transparent plastic film mulching (PM) and no mulching (CK). We researched the dynamic feature of soil organic carbon (SOC) and its components at different growth stages of winter wheat under different mulching methods, including microbial biomass carbon (MBC), dissolved organic carbon (DOC), mineralizable carbon (PCM) and easily oxidized carbon (EOC). The results showed that SOC and its components in different soil layers decreased with an increase of soil layer depth. Compared with CK, the SM treatment increased SOC in the 0–40 cm soil layer by 2.83–8.92%, MBC by 12.09–18.40%, DOC by 3.73–9.79%, PCM by 4.82–12.48% and EOC by 6.01–11.68% during the different periods, and an overall increase was greater in the 0–20 cm soil layer than in the 20–40 cm soil layer. The impact of PM on SOC was less significant compared with CK; MBC and PCM had an overall positive effect, and DOC and EOC behaved differently from year to year. In conclusion, SM can improve SOC and its components content in dryland of northwest China, which is conducive to the sustainable management and efficient utilization of maize straw resources and has significant ecological benefits.

To investigate the feasibility of replacing plastic film with straw in semi-arid areas and establishing coordinated cultivation technology for high winter wheat yield and efficient resource utilization, a two-year field experiment was conducted under six treatments, specifically CK (no-mulching), S1 (59% of the field area straw mulched), S2 (50% of the field area straw mulched), S3 (42% of the field area straw mulched), BM (full-cover transparent plastic mulch), and HM (full-cover black plastic mulch). The effects of mulching measures on soil moisture, water consumption characteristics, yield and water use efficiency (WUE) of winter wheat farmland in rain-fed semiarid regions were studied. The results showed that, compared with CK, straw strip mulching reduced total water consumption by 15.39 mm on average, the soil organic carbon content at the 0–40 cm soil layer increased by 4.68%, yield by 6.90%, WUE_r by 11.27%, and WUE_b by 16.51%. Compared with CK, the total water consumption and soil organic carbon content in each growth period of plastic film mulching were not significantly different, but the yield, WUE_r, and WUE_b increased by 16.28%, 15.29%, and 25.50%, respectively. Among the three straw strip mulching treatments, treatment S3 had the highest yield, which was equivalent to that of plastic film mulching. The S3 treatment with 42% of the field area straw mulched is recommended in this stusy as the optimal replacement of plastic film mulching in semi-arid environments.

Hyperspectral remote sensing provides a powerful tool for crop nutrient monitoring and precision fertilization, yet its application is hindered by high-dimensional redundancy and inter-band collinearity. This study aimed to improve maize nitrogen estimation by constructing three types of two-dimensional full-band spectral indices—Difference Index (DI), Simple Ratio Index (SRI), and Normalized Difference Index (NDI)—combined with spectral preprocessing methods (raw spectra (RAW), first-order derivative (FD), and second-order derivative (SD)). To optimize feature selection, three strategies were evaluated: Grey Relational Analysis (GRA), Pearson Correlation Coefficient (PCC), and Variable Importance in Projection (VIP). These indices were then integrated into machine learning models, including Backpropagation Neural Network (BP), Random Forest (RF), and Support Vector Regression (SVR). Results revealed that spectral index optimization substantially enhanced model performance. NDI consistently demonstrated robustness, achieving the highest grey relational degree (0.9077) under second-derivative preprocessing and improving BP model predictions. PCC-selected features showed superior adaptability in the RF model, yielding the highest test accuracy under raw spectral input (R² = 0.769, RMSE = 0.0018). VIP proved most effective for SVR, with the optimal SD–VIP–SVR combination attaining the best predictive performance (test R² = 0.7593, RMSE = 0.0024). Compared with full-spectrum input, spectral index optimization effectively reduced collinearity and overfitting, improving both reliability and generalization. Spectral index optimization significantly improved inversion accuracy. Among the tested pipelines, RAW-PCC-RF demonstrated robust stability across datasets, while SD-VIP-SVR achieved the highest overall validation accuracy (R² = 0.7593, RMSE = 0.0024). These results highlight the complementary roles of stability and accuracy in defining the optimal pipeline for maize nitrogen inversion. This study highlights the pivotal role of spectral index optimization in hyperspectral inversion of maize nitrogen content. The proposed framework provides a reliable methodological basis for non-destructive nitrogen monitoring, with broad implications for precision agriculture and sustainable nutrient management.

Gentiana apiata N. E. Brown (Gentianaceae) is a perennial herb plant and only grows in Qinba Mountains in China. Here, we first characterized the complete nucleotide sequence of chloroplast (cp) genome of G. apiata via Illumina next generation sequencing platform. The complete chloroplast genome of G. apiata was 144,274 bp in length, comprising of a large single copy (LSC) region of 77,353 bp, a small single copy (SSC) region of 17,009 bp, and two inverted repeat regions (IRs) of 24,956 bp. The cp genome contains 127 genes, including 82 protein-coding genes, 35 tRNA, eight rRNA genes, and two pseudogenes. Phylogenetic analysis based on 18 cp genome sequences showed that G. apiata closely related to congeneric species.

Against the background of increasingly scarce water resources and intensifying water use conflicts, achieving the scientific and optimized allocation of water resources has become crucial to ensuring regional sustainable development. Based on the traditional water resource optimization models that consider social, economic, and ecological objectives, this study introduces a spatial equilibrium level as a fourth optimization objective, constructing a multi-objective water resource allocation optimization model. The model simultaneously incorporates constraints on water supply, water demand, and decision variable non-negativity, as well as coupling coordination constraints among the water resources, socio-economic, and ecological subsystems within each water use unit. The NSGA-III algorithm is employed to obtain the Pareto front solution set for the four objectives, followed by a comprehensive ranking of the Pareto solutions using an entropy-weighted TOPSIS method. The solution exhibiting the best trade-off among the four objectives is selected as the decision basis for the water allocation scheme. Taking Jiuquan, Jiayuguan, and Zhangye cities in western Gansu Province as the study area, the results indicate that the optimal allocation scheme can guide the cities to shift from “water-deficit usage” toward “water-saving usage,” achieving a reasonable balance between meeting water demand and water conservation requirements. This promotes coordinated development among the water resource, socio-economic, and ecological subsystems within each city as well as among the cities themselves, thereby facilitating sustainable utilization of water resources and sustainable development of socio-economics and the ecological environment. The findings can serve as a reference for water resource allocation strategies in the study region.