Cucumber green mottle mosaic virus (CGMMV) is one of the important plant viruses that harm melon crops, and its movement protein (MP) plays a key role in virus transmission and pathogenicity. This study cloned the MP gene of CGMMV through RT-PCR, constructed the prokaryotic expression vector pET-32a-MP, and successfully induced the expression of a recombinant MP fusion protein with a molecular weight of approximately 48 kDa in Escherichia coli BL21 (DE3). Purification of high purity protein (purity degree≥80%) using nickel column affinity chromatography was used to prepare polyclonal antibodies against the New Southwest White Rabbit. Western blot and ELISA analysis showed that the titer of the anti serum reached 409 600, and it could specifically recognize the MP protein in CGMMV infected leaves, without cross reactivity with other viruses such as Tobacco mosaic virus (TMV) and Watermelon mosaic virus (WMV). The study would provide important tools for rapid virus detection, immunohistochemistry, and protein function research, which is of great significance for ensuring the safe production of melon crops.

Rubber plantations represent one of the most widespread planted forest ecosystems in tropical regions, playing a significant role in regional carbon cycles. Global climate change has led to increasingly frequent Compound Drought-hot Events (CDHEs), posing severe challenges to rubber plantation ecosystem functions. This study focuses on rubber plantations in Hainan Island, China. Utilizing standardized soil moisture and land surface temperature indices, we constructed a Compound Drought-Hot Index (CDHI) based on Copula functions to analyze the spatiotemporal dynamics of CDHEs from 2001 to 2020, validated against historical typical drought and hot events in Hainan. Using Gross Primary Productivity (GPP) as an evaluation metric, Copula-based conditional probabilities were applied to quantitatively assess GPP loss risks under different drought-hot scenarios and evaluate compound effects. Key findings include: (1) The developed CDHI accurately identified severe CDHEs during 2001—2020, with 2005 and 2020 being the most extreme years. (2) Across all intensity levels (mild, moderate, severe), CDHEs caused significantly higher probabilities of GPP loss in rubber plantations compared to isolated drought or hot events (P<0.001), confirming the synergistic amplification effect of compound events. (3) GPP loss risks exhibited distinct spatial heterogeneity influenced by regional topography-climate interactions, with northwestern rubber plantation areas showing consistently higher vulnerability across all scenarios. (4) CDHE impacts demonstrated marked seasonal differences, being substantially greater during the dry season (particularly April—May) than the rainy season, coinciding with critical phenological stages of rubber trees and forming sensitive windows of environmental stress. This study reveals the spatiotemporal differentiation of CDHE impacts on rubber plantation GPP, providing scientific foundations for regional climate-resilient management strategies and sustainable rubber plantation practices. It also offers methodological references for studying extreme climate responses in other tropical planted forest ecosystems.

The study was aimed to develop organic fertilizers replacing chemical fertilizers and reveal the effect mechanism of fertilization on plant health, endophytic bacterial community structures in tomato roots. Five fertilization treatments were set up, no fertilization (CK), application with chemical fertilizer (T₁), peanut cake fertilizer (T₂), soybean cake fertilizer (T₃) and tea cake fertilizer (T₄) at the identical nitrogen level. Meanwhile, based on high-throughput sequencing technology, the endophytic bacterial community structure in tomato roots was analyzed. The numbers of endophytic bacterial communities in tomato roots at different taxonomic categories were all improved over CK. Meanwhile, except T₄ treatment, the diversity and richness of endophytic bacteria community in tomato roots under T₂ and T₃ were all higher than those of T₁. At the phylum or genus level, the endophytic bacterial composition and the percentage in tomato roots were all altered by different fertilizer treatments. Among them, not only the numbers of dominant endophytic bacterial were improved, but also the endophytic bacterial community structures were reshaped. Particularly, some beneficial bacterial genera, such as Rhizobium, Bacillus, Microbacterium, Castellaniella, with the function of growth promotion, plant resistance and quality improvement enriched in tomato roots under T₂ treatment. Compared with CK, the diversity and richness of endophytic bacterial community in tomato roots were improved, and the endophytic bacterial composition in tomato roots could be reshaped by different fertilization. The highest diversity and richness of endophytic bacterial communities could be found in tomato roots under T₂ treatment, followed by T₃ treatment. The beneficial endophytic bacterial genera, such as Rhizobium, Bacillus, Microbacterium and Castellaniella enriched in tomato roots under T₂ treatment only. It indicated that the effect of peanut cake fertilizer on plant health was better than those of chemical fertilizer, soybean and tea cake fertilizers under the identical nitrogen levels.

It is necessary to establish a sensitive, selective, efficient and accurate method for the simultaneous detection of multiple antibiotic residues in soil. In this study, the method of solid phase extraction-high performance liquid chromatography-tandem mass spectrometry (LC-MS/MS) for the determination of 20 antibiotics simultaneously in soil was established by optimizing the extraction conditions (extractant type, extraction method, shading and overnight) and the purification conditions (cartridge type). Under the optimized analysis conditions, the target antibiotics showed good linearity in the linear range of 0.10-100.00 μg/kg, with R² ranging from 0.9951-0.9999. The recovery ratio ranged from 62%-116%, the relative standard deviation was 0.9%-16.8%, the limit of detection was 0.01-0.50 μg/kg, and the limit of quantification was 0.04-2.00 μg/kg. The method was applied for the detection of antibiotic residues in soil samples collected from greenhouse vegetable bases in Wuqing district, Tianjin, China. It showed that 4 classes of antibiotics were detected in the soil, and up to 16 of target antibiotics could be detected simultaneously. The total residual concentration of antibiotics in the soil varied from 2.52-142.00 μg/kg. The highest concentration of individual antibiotic was 105.00 μg/kg (oxytetracycline). The method established in the study is aim to provide reference for the formulation of the standard for the simultaneous detection of multiple antibiotic residues in soil and provide methodological basis for the prevention and control of emerging contaminants.

The study of methane (CH₄) emission flux from paddy fields is an important part and hot topic in the current study of carbon cycle in terrestrial ecosystems. The multi-time scale characteristics of CH₄ emission flux and its influencing factors [gross primary productivity (GPP), latent heat flux (LE), air temperature (TA), soil temperature (TS)] in paddy fields were analyzed based on wavelet analysis and eddy covariance flux observation data. The CH₄ emission flux and its influencing factors of paddy field ecosystems had obvious seasonal variation characteristics. The CH₄ emission flux from paddy fields in the rice growing season showed obvious single-peak diurnal variation characteristics, while the CH₄ emission flux from paddy fields in the non-rice growing season was low, and the diurnal variation characteristics were not obvious. GPP, LE, TA and TS showed obvious single-peak diurnal variation patterns. Based on continuous wavelet transform and wavelet variance curve analysis, CH₄ flux, GPP, LE, TA and TS had an obvious 1-day cycle during the rice growing season (July—November), while GPP, LE, TA and TS also had a half-day cycle, and TA and TS also had a long-term cycle of 4 days. From the cross wavelet transform and wavelet coherence spectrum analysis, the resonance period between CH₄ flux and GPP, LE, TA and TS was about 1 day. There was a significant resonance relationship in this frequency domain, and oscillation cohesion and coherence were the strongest. However, in the high-frequency band area, there was also a secondary resonance period of about 0.5 days between CH₄ flux and GPP, LE, TA and TS. In addition, in other frequency bands, the resonance energy of CH₄ flux and GPP, LE, TA and TS was low, but in the resonance period of 4 days and 7-10 days, CH₄ flux had good coherence with TA and TS from mid-August to mid-October. This study analyzed the multi-time scale characteristics of CH₄ flux and its influencing factors from paddy fields and would provide a scientific reference for studying the quantitative relationship between CH₄ flux and its driving factors at different time scales.

The study prepared slow-release fertilizers with different ratios of zeolite and nano-zeolite and compared them with conventional urea fertilizers to study the effect of zeolite and nano-zeolite as urea carriers on soil nutrient leaching, including nutrient release patterns, leaching characteristics, and impacts on soil nutrient content. A soil column leaching experiment was conducted with seven treatments, conventional urea (CK) and slow-release fertilizers with different ratios of zeolite (ZU₁, ZU₂, ZU₃) and nano-zeolite (nZU₁, nZU₂, nZU₃). The study analyzed the dynamic changes and cumulative leaching amounts of ammonium nitrogen, nitrate nitrogen, total nitrogen, and total phosphorus in the leachate and measured the soil nutrient content after leaching. Zeolite and nano-zeolite-based slow-release fertilizers significantly prolonged fertilizer effectiveness and reduced the concentration of ammonium nitrogen, nitrate nitrogen, total nitrogen, and total phosphorus in the leachate (P<0.05). Treatments with 50% zeolite (ZU₃) and 50% nano-zeolite (nZU₃) were the most effective in reducing nutrient leaching. Specifically, ZU3 reduced ammonium nitrogen, nitrate nitrogen, total nitrogen, and total phosphorus in the leachate by 7.19%, 6.54%, 14.51% and 9.72%, respectively, while nZU₃ reduced them by 7.20%, 6.91%, 12.84% and 6.36%, respectively. The slow-release fertilizers also significantly increased soil nitrogen content, with total nitrogen in ZU₃ and nZU₃-treated soils increasing by 14.88% and 19.83% compared to the control. From the perspective of nutrient release patterns, it was observed that zeolite and nano-zeolite slow-release fertilizers exhibited a lower release rate in the early stages of the experiment. This is attributed to the strong adsorption properties of zeolite and nano-zeolite, which can effectively slow down the release of urea. The release peak of the fertilizers was delayed to 6–8 days, and the release became more steady in the later stages of leaching. This pattern suggests that the application of zeolite and nano-zeolite can effectively reduce the risk of nutrient loss due to leaching, thereby improving the utilization efficiency of fertilizers. Furthermore, the comparison between zeolite-based and nano-zeolite-based slow-release fertilizers revealed that the differences in reducing the leaching of ammonium nitrogen, total nitrogen, and total phosphorus were not obvious, but both of them had significantly better performance than conventional urea in terms of reducing nutrient leaching (P<0.05). In summary, the use of zeolite and nano-zeolite as urea carriers in slow-release fertilizers showed to be an effective strategy for delaying nutrient release, reducing nitrogen and phosphorus leaching, and enhancing soil nutrient content. The optimal results were achieved with a 50% zeolite and 50% nano-zeolite ratio, indicating a promising application potential in agricultural practices. This approach not only contributes to the sustainable management of soil fertility but also offers a potential solution for reducing the environmental impact associated with excessive fertilizer use.

This study aimed to identify the key factors influencing vegetation changes on Hainan Island under Typhoon Yagi, determine the disaster thresholds, and provide scientific support for typhoon disaster prediction, assessment, and mitigation. Using the GEE platform, Sentinel-2 NDVI data from August 19 to September 18, 2024, covering the period before and after Typhoon Yagi landfall, were analyzed in conjunction with topographic and meteorological variables (such as typhoon path, maximum wind speed, minimum pressure, and precipitation). An XGBoost model was applied to predict post-typhoon NDVI changes, and SHAP analysis was used to explore the nonlinear effects of meteorological factors. The XGBoost model yielded a coefficient of determination (R²) of 0.75 and a root mean square error (RMSE) of 0.12, indicating a high prediction accuracy for post-typhoon NDVI changes. Feature importance analysis within the XGBoost model revealed that pre-typhoon NDVI, typhoon distance, and maximum wind speed were the primary influencing factors, while slope and aspect had a weaker effect. Further SHAP analysis confirmed that typhoon distance and maximum wind speed were the key meteorological factors driving NDVI changes. When the typhoon distance was less than 76.4 km, NDVI showed a significant decline, while NDVI changes intensified when the maximum wind speed exceeded 29.3 m/s, highlighting the crucial role of strong winds in vegetation damage. Regions within 76.4 km (approximately within the 11th wind radius) of the typhoon's path and with maximum wind speeds greater than 29.3 m/s experienced severe vegetation damage. It is recommended that the high-risk areas be prioritized in typhoon warnings and pre-disaster defense measures.

Tropical agriculture is critical to global food security and rural revitalization. However, its modernization is constrained by complex terrain, climatic variability and frequent disasters. Remote sensing, with its capacity for large-scale, real-time information acquisition, has emerged as a key enabler for smart and precision agriculture in tropical regions. This study established a framework for remote sensing research in China's tropical agriculture, proposing an integrated “space-air-ground” observation strategy and explaining the resolution-to-application matching mechanism, in which low-, medium-, and high-resolution observations are respectively suited for regional resource surveys, farmland growth monitoring, and field-scale phenotypic diagnostics. It reviewed progress in natural resource surveys, crop monitoring yield estimation, disaster assessment and ecological evaluation, analyzed challenges such as insufficient ground-truth data, limited model transferability, poor regional adaptability and low transformation efficiency; and proposes key innovation paths including coordinated sensing, multi-modal data fusion, intelligent modeling and digital twin-based forecasting. Finally, it outlined future directions focusing on system integration, intelligent perception, and global collaboration. The findings would provide theoretical and technical support for enhancing remote sensing applications in tropical agriculture.

The yield of tropical crops is highly sensitive to climate conditions, and accurately modeling the meteorological-driven mechanisms is crucial for improving tropical agricultural productivity and climate adaptability. This study systematically compared the prediction performance of six machine learning models, including LGBM, RF, XGBoost, AdaBoost, SVM and MLR based on natural rubber, mango, pineapple and banana in Hainan. The SHAP method was used to quantify the contribution and non-linear response characteristics of meteorological factors. The LGBM model demonstrated the best prediction performance, with an average R² of 0.945 for the test set (the R² of rubber, mango, pineapple and banana were 0.942, 0.902, 0.954 and 0.983, respectively), and average RMSE and MAE of 1.436 t/hm² and 1.150 t/hm², significantly outperforming the other models (the R² of RF, XGBoost, AdaBoost, SVM, MLR were 0.773, 0.563, 0.589, 0.368 and 0.508, respectively). The meteorological-driven mechanisms exhibited significant crop-specific differences. Rubber yield was mainly driven by solar radiation (the contribution was 14.7%) and temperature factors (the contribution of monthly minimum temperature and monthly maximum temperature were 14.4% and 11.7%, respectively). Mango yield was highly sensitive to monthly maximum temperature (the contribution was 19.0%) and vapor pressure deficit (the contribution was 18.5%). Pineapple and banana yield were dominated by soil moisture (the contribution was 18.9%) and relative humidity (the contribution was 23.6%), respectively. Based on the findings, differentiated agronomic management recommendations for each crop type were proposed. This study demonstrates that machine learning, combined with explainability methods, can effectively elucidate the climate response mechanisms of tropical crops, providing theoretical support for regional agricultural precision management.

Transcriptional regulation is one of the pivotal mechanisms governing biological activities. Natural rubber (NR), a crucial industrial raw material, is predominantly derived from rubber trees. The transcriptional regulation of NR biosynthesis has emerged as a focal point in the fundamental research of rubber tree in recent years. Mevalonate diphosphate decarboxylase (MVD) is a key enzyme in the biosynthetic pathway of NR, yet its transcriptional regulation remains poorly understood. In this study, a 1500 bp promoter region of the HbMVD1 gene from rubber tree was cloned. Sequence analysis revealed the presence of various cis-acting elements, including the responsive to jasmonic acid, ethylene, mechanical wounding, light, and MYC binding. A gradient assay for autoactivation suppression demonstrated that the transcriptional autoactivation of the pAbAi-pHbMVD1 bait vector was effectively inhibited at an AbA concentration of 100 ng/mL. 13 non-redundant proteins from a rubber tree latex yeast cDNA library that interact with the HbMVD1 promoter, including one WRKY transcription factor (HbWRKY1), one heat shock transcription factor, and one G-box binding factor were identified utilizing yeast one-hybrid screening. Tissue-specific expression analysis indicated that the genes encoding the proteins exhibited distinct expression patterns across different tissues. Transcriptional regulation analysis showed that HbWRKY1 activated the expression of HbMVD1 in vivo. Molecular docking revealed that HbWRKY1 bound to the mechanical wounding element of the HbMVD1 promoter, and 11 potential amino acid binding sites were identified. This study would lay the groundwork for further elucidating the transcriptional regulatory network of HbMVD1 in rubber tree.