This study integrated airborne LiDAR point cloud features with a hyperparameter optimization algorithm to develop a precise carbon stock estimation model for Pinus elliottii plantations in wetlands. The aim is to provide a scientific basis for large-scale carbon sink monitoring, thereby supporting forest management decisions that align with the "dual carbon" goals.

Taking Pinus elliottii plantation in Guangdong Province as the research object, 101 point cloud feature variables were extracted based on airborne LiDAR data. The recursive feature elimination combined with cross-validation (RFE-CV) was used to screen the optimal feature combination, and the Optuna hyperparameter optimization framework based on the TPE algorithm (with RMSE as the objective function) was introduced to construct random forest (RF), k-nearest neighbor (KNN) and extreme gradient boosting (XGBoost) models. The model performance was evaluated by R², RMSE and MAE, and the SHAP value was used to analyze the contribution mechanism of features to the prediction of carbon storage of Pinus elliottii.

1) Five optimal feature variables (A_IH40, A_IH80, H₁₀, H₂₅, I_max) closely related to the estimation of Pinus elliottii carbon storage were selected by 5-fold RFE-CV. Compared with the model using all variables, feature selection significantly improved the prediction accuracy of each model. Among them, the XGBoost model has the most obvious effect (R² increased from 0.78 to 0.89, RMSE decreased from 0.63 to 0.46 t, MAE decreased from 0.48 to 0.41 t); 2) The Optuna hyperparameter optimization framework based on the TPE algorithm is significantly better than random search: the RMSE value of the XGBoost model is reduced by 2.31%, and the error adjustment percentage (13.13%) is 3.57 times that of random search. The convergence iterations of the RF and KNN models are reduced by 33.33% and 49.18%, respectively; 3) After optimization by Optuna-TPE, the accuracy of each model has been significantly improved: the XGBoost model achieves the highest accuracy, with its R² significantly increasing from 0.89 to 0.93, thus becoming the optimal model; 4) SHAP value analysis revealed the difference of feature dependence in different slash pine models: A_IH40 dominated XGBoost prediction (SHAP = 0.64), while H₂₅ was more critical to the RF model (SHAP = 0.28).

This study showed that feature selection and Optuna hyperparameter tuning based on the TPE algorithm are crucial to improve the performance of the Pinus elliottii carbon storage estimation model. The optimized XGBoost model combined with the optimal feature combination realized the high-precision estimation of the carbon storage of Pinus elliottii plantations.

To investigate the effects of different nitrogen (N), phosphorus (P), and potassium (K) fertilization ratios on the accumulation of nitrogen metabolites, soluble sugar content, vegetative growth, and reproductive organ development in 19-year-old Taiwania cryptomerioides. The goal was to identify an optimal N-P-K fertilization scheme to promote its growth and development.

An L₁₆(4³) orthogonal experimental design with three factors (N, P, K) and four levels was employed. The contents of soluble sugar, soluble protein, and total nitrogen were determined in the buds, flowers, and fruits. Additionally, growth parameters of vegetative and reproductive shoots were measured.

1) A comprehensive evaluation indicates that the N3P4K2 treatment has the most significant overall promotional effect on the main agronomic traits. Under this treatment, bud soluble protein content reaches 9.84 mg/g and total nitrogen content is 1.97%, representing increases of 116.26% and 28.76% over the control, respectively. The dry weights of vegetative and reproductive shoots reach 1.21 g and 1.38 g, which are 55.13% and 2.22% higher than the control. Bud soluble sugar content under this treatment is 49.05 mg/g, a 54.68% increase. The N3P2K4 treatment also shows excellent performance, ranking second overall, with increases of 114.95% in bud soluble protein, 26.14% in bud total nitrogen, 47.68% in vegetative shoot dry weight, 1.48% in reproductive shoot dry weight, and 48.50% in bud soluble sugar compared to the control; 2) Nitrogen exerts an extremely significant promotional effect (P<0.01) on soluble protein content and vegetative growth indicators, with effects typically peaking at the N3 level and showing a downward trend at the N4 level. In contrast, total nitrogen content is generally highest at the N4 level; 3) Phosphorus also shows an extremely significant promotional effect (P<0.01) on most measured indicators, with the P3 level performing best for protein and total nitrogen accumulation; 4) The main effect of potassium is not significant (P>0.05) for most metabolic and growth indicators; however, its presence at specific levels is an essential component of the optimal treatments, suggesting a crucial role in nutrient balance and interaction.

A balanced N-P-K application, particularly N3P4K2 and N3P2K4 can significantly optimize nitrogen metabolism and soluble sugar accumulation in Taiwania cryptomerioides.

To breed novel Camellia oleifera cultivars characterized by spring flowering, superior fruit economic traits, and excellent oil quality, this study conducted a comprehensive evaluation of distant hybridization progeny, aiming to screen elite spring-flowering germplasm resources.

Twenty-seven initially selected spring-flowering superior individuals from the hybrid progeny of ‘Youza 2’ (YZ2) × ‘Deyou 2’ (DY2) served as experimental materials. Systematic observations were made on their flowering phenology, floral organ morphology, pollen viability, and fruit economic traits. Data were analyzed using correlation analysis, difference analysis, and principal component analysis.

Significant segregation in flowering traits was observed among the 27 spring-flowering Camellia oleifera superior individuals. The earliest initial flowering date occurred in YD21, while the latest was recorded for YD11. YD26 exhibited the longest full-bloom period, lasting 32 days. Floral organ traits showed varying degrees of variation among individuals, with style number ranging from 3 to 5 lobes and petal number ranging from 5 to 10. Notably, YD22 displayed the largest flower diameter (89.56 mm), consistent with entomophilous characteristics. Correlation analysis revealed highly significant positive correlations (P<0.01) between petal number and style number, and between flower diameter and style length, indicating coordinated regulation in floral organ development. Pollen viability varied markedly among individuals, ranging from 35.35% to 83.31%. YD17 demonstrated the highest pollen viability (83.31%), and the phenomenon of “pseudo-pollen” was observed. Regarding fruit traits, YD6 possessed the thinnest pericarp (1.32 mm), with its fresh seed yield rate (52.99%) and kernel oil content (51.17%) being significantly superior to parental performance. YD17 exhibited the highest fresh seed yield rate (63.31%), standing out prominently among the population. Unsaturated fatty acids accounted for 85.40%-88.01% of the total fatty acids in the oil. YD11 displayed the highest oleic acid content (79.77%).

Comprehensive evaluation identified superior individuals YD3, YD7, and YD17 as outstanding performers for key economic traits, including fruit size, seed yield rate, and oleic acid content. These individuals represent valuable germplasm resources for further breeding and research.

To reveal the variation patterns and driving mechanisms underlying leaf phenotypic diversity in Quercus mongolica and provide a data basis for constructing Quercus mongolica germplasm banks, breed suitable provenances, and to promote sustainable utilization.

Thirteen leaf phenotypic traits were systematically measured for 55 Quercus mongolica provenances from northeast and north China. Nested analysis of variance was used to quantify variation contributions among and within provenances; phenotypic differentiation coefficients were employed to analyze the degree of genetic differentiation; principal component analysis (PCA) and UPGMA clustering were combined to explore the classification value of traits; regression models were utilized to evaluate the impact of climatic differences (e.g., annual average temperature, seasonal rainfall) between provenance and planting sites on phenotypes.

All 13 traits show significant differences among provenances (P<0.05), with coefficients of variation (CV) ranging from 18.81% to 63.50%. Among them, leaf area (CV=63.50%), petiole length (45.42%), and leaf base angle (39.86%) exhibit the most significant variations. The average phenotypic differentiation coefficient is 61.99%, and the variance contribution among provenances (57.65%) is significantly higher than that within provenances (35.09%), indicating that variation mainly originates from geographical population differentiation. Correlation analysis shows that different leaf traits are mainly highly correlated with leaf length and leaf width. PCA extracts 3 principal components (cumulative contribution rate 66.60%), with dominant factors being leaf length (PC1 loading 0.954), leaf tip angle (PC2 loading 0.745), and leaf base angle (PC3 loading 0.743) in sequence. Cluster analysis divides the 55 provenances into 3 types: large-leaf type (6 provenances), medium-leaf type (23 provenances), and small-leaf type (25 provenances). The medium-leaf and small-leaf types account for 90.9%, with classification based on leaf size (leaf length, leaf width) and angle characteristics (leaf tip angle, leaf base angle). Climatic differences between provenance and planting sites have significant effects on 6 traits (P<0.05), but the explanatory rates (coefficient of determination, R²) are generally low (<0.17).

Leaf phenotypic diversity in Quercus mongolica is dominated by genetic differentiation among provenances. Leaf area, petiole length, leaf base angle, leaf perimeter, and leaf vein angle are key indicators of phenotypic differentiation; leaf length, leaf tip angle, and leaf base angle can serve as core indices for germplasm classification. Climatic factors have a weak impact on traits, indicating that genetic background is the main cause of variation.

To investigate the effects of different fertilization treatments on the growth and physiology of Phoebe bournei seedlings, and to provide a theoretical basis for developing scientific fertilization strategies to cultivate high-quality seedlings that are both tall and robust, thereby improving afforestation survival rates.

Two-year-old Phoebe bournei seedlings were used as experimental materials. A factorial design was employed with four compound fertilizer levels (0, 4, 8, 16 g per seedling) and four bio-fertilizer levels (0, 5, 10, 20 g per seedling), forming 16 treatment combinations. Seedling height, ground diameter, biomass, chlorophyll content, soluble protein content, as well as soluble sugar and starch contents in various plant parts (leaves, branches, roots) were measured. Multiple comparisons and correlation analysis were used to systematically study the effects of fertilization on seedling growth and NSC accumulation and allocation.

Fertilization significantly promoted the growth of Phoebe bournei seedlings. Treatments T6 (4 g compound fertilizer + 10 g bio-fertilizer) and T7 (4 g compound fertilizer + 20 g bio-fertilizer) showed the most pronounced increases in seedling height, ground diameter, and biomass. The medium fertilizer group overall exhibited the best comprehensive promotion effect, resulting in seedlings with suitable height-to-diameter and shoot-to-root ratios, while maintaining high seedling quality indices and overall growth coordination. Therefore, the combined application of 4-8 g compound fertilizer combined with 10-20 g biofertilizer is recommended to achieve both quality improvement and efficient NSC accumulation in Phoebe bournei seedlings. Correlation analysis indicated widespread correlations among seedling height, ground diameter, soluble protein, chlorophyll, biomass, and NSC contents in different parts. Main effect analysis revealed that compound fertilizer had a strong main effect, enhancing most indicators including seedling height, ground diameter, total biomass, and seedling quality index. Bio-fertilizer had a relatively weaker main effect but was particularly prominent in promoting NSC accumulation in roots and branches.

The combined application of compound fertilizer and biofertilizer is recommended to balance rapid growth and NSC accumulation in Phoebe bournei seedlings, shorten the growth cycle, and enhance stress resistance. This study provides support for supplying high-quality seedlings for the establishment of Phoebe bournei plantations and the restoration of wild populations.

Exploring the physiological response to drought, to screen drought-resistant varieties, accelerate the breeding of drought-resistant Phalaenopsis varieties, meet the market demand in arid areas, and improve economic benefits.

Eleven Phalaenopsis cultivars were selected, and their photosynthetic parameters, chlorophyll fluorescence characteristics and physiological indexes were investigated after natural drought following watering, and the drought-resistant varieties were screened out by using membership functions and cluster analysis.

As the water content decreases, the net photosynthetic rate, chlorophyll fluorescence parameters, and chlorophyll content of most varieties continuously decline, while the intercellular CO₂ concentration, electrical conductivity, soluble sugar content, and malondialdehyde content continuously increase. The contents of CAT, SOD, and soluble proteins show a trend of first increasing and then decreasing. The drought tolerance evaluation system constructed based on the membership function method indicates that the drought tolerance ranking of the varieties is as follows: ‘Gold Princess’ > ‘Dior’ > ‘Sunrise’ > ‘Aurea Mocha’ > ‘Las Vegas’ > ‘Anna’ > ‘Prince Berry’ > ‘Bordeaux’ > ‘Gongxi Facai’ > ‘Man Tianhong’ > ‘Dream Lady’. Through cluster analysis, the 11 Phalaenopsis varieties are divided into two categories based on their drought tolerance. The first category consists of six drought-tolerant varieties, namely ‘Gold Princess’ ‘Dior’ ‘Sunrise’ ‘Aurea Mocha’ ‘Las Vegas’ and ‘Anna’, the second category includes five general varieties, namely ‘Prince Berry’ ‘Bordeaux’ ‘Gongxi Facai’ ‘Man Tianhong’ and ‘Dream Lady’.

‘Gold Princess’ ‘Dior’ ‘Sunrise’ ‘Aurea Mocha’ ‘Las Vegas’ and ‘Anna’ can adapt to the dry climate of Ningxia and are suitable for planting and popularizing in Ningxia; ‘Prince Berry’ ‘Bordeaux’ ‘Gongxi Facai’ ‘Man Tianhong’ and ‘Dream Lady’ may need frequent watering in the arid area of Ningxia and the cost is high, so they should be planted appropriately.

To explore the differential mechanisms of carbohydrates in dioecious plants, this study took Fraxinus velutina as the research object. It aimed to analyze the differences in carbohydrate allocation patterns between male and female plants at different developmental stages, thereby providing a scientific basis for the regulation of carbon metabolism in sexually dimorphic plants and the management of urban forests.

Specimens were gathered during the fruiting phase (S1), fruit expansion period (S2), and fruit ripening stage (S3). Branches in the upper southern section of the trees were tested to determine the levels of non-structural and structural carbohydrates.

Growth stages and their combined effect with gender notably affect structural carbohydrates and some non-structural carbohydrates (P<0.05). In both genders of F. velutina, leaf cellulose content first increases and then decreases, with females showing significantly lower levels than males at S1. Lignin content declines steadily, and females have lower levels in all stages, with a marked difference at S2 (P<0.05). Regarding non-structural carbohydrates: female starch content drops dramatically from S1 to S2 and is significantly lower than that of males at S2 (P<0.05); fructose content first decreases and then increases, with females were significantly lower at S2; sucrose content follows a similar trend, with females showing significantly lower levels at S2 (P<0.05). Correlation analysis indicates that in females, cellulose is mostly negatively associated with non-structural carbohydrates (such as starch and soluble sugars), while in males, starch is mostly positively associated with soluble sugars and their components. This reflects gender-specific coordination in carbon metabolism and trade-off strategies in carbon allocation.

Owing to distinct life history strategies, male and female F. velutina exhibit differences in carbohydrate allocation during development: males retain higher levels of structural carbohydrates in leaves, relying on the advantages of cellulose and lignin to support vegetative growth; females display more adaptable mobilization of non-structural carbohydrates, adjusting starch and soluble sugars dynamically to meet reproductive needs.

To explore the water sources of riparian forests and their response to the water level of the Yellow River in the Ulan Buh desert section of the Yellow River basin, reveal the water use strategies of Salix matsudana, understand their adaptability under drought conditions, and to provide a theoretical basis for ecological protection and restoration in the desert area of the Yellow River basin.

Salix matsudana was selected from the river bank of the Ulan Buh desert of Yellow River “Jizi bend”. In the growing season of 2022-2023, plant stems were collected, as well as forest soil, groundwater, Yellow River water and other potential water sources of plants. Water was extracted from plant and soil samples by a freeze-vacuum water extraction device, and the stable hydrogen and oxygen isotopes of each water sample were determined and analyzed. The contribution ratio of different potential water sources to plants was calculated using MixSIAR model.

The water source of Salix matsudana had obvious seasonal changes. In May 2022, Salix matsudana mainly absorbed deep soil water, accounting for 84.1%. In June, July, August and September, the absorption proportion of deep soil water decreased (44.6%, 36.9%, 26.0%, 54.1%). Concurrently, the proportion of water absorbed from the middle soil increased by 24.3%, 13.7%, 8.6% and 7.2%, respectively. Additionally, the proportion of groundwater absorption increased by 6.9%, 21.5%, 38.3% and 19.1% respectively during these months. From May to September 2023, Salix matsudana mainly absorbed deep soil water and groundwater. The proportion of deep soil water absorption was 65.6%, 38.9%, 63.3%, 52.3% and 30.2%, and the proportion of groundwater absorption was 14.3%, 31.6%, 18.6%, 29.1% and 48.2%, respectively. There was no significant difference between 2022 and 2023 in the contribution ratio of different potential water sources (P>0.05). The proportion of groundwater absorbed by Salix matsudana increased with the water level of the Yellow River, and the proportion of Yellow River water to the water source of Salix matsudana decreased.

The main water source of Salix matsudana is relatively reliable deep soil water and groundwater. The water level of the Yellow River will affect the water source of Salix matsudana to a certain extent, especially the absorption ratio of groundwater and Yellow River water, which reflects the adaptability of Salix matsudana to drought.

To reveal the spatial differentiation law of the ecological service value of typical sand-land governance modes in the middle reaches of the Heihe river, the characteristics of the adaptability of the six sand-land governance modes, and to provide a scientific basis for the optimization of the models and spatial optimization allocation of regional sand-land governance, and to promote the maximization of ecological restoration benefits and regional sustainable development.

This study analyzed the spatial differentiation characteristics of different governance modes through the processes of questionnaire survey, expert scoring, literature review, field investigation, indoor detection and data analysis. A multi-level ecological service value assessment mode for governance patterns was constructed based on the principles of the Analytic Hierarchy Process (AHP) to determine the weights of evaluation indicators. According to the weighted quantitative scoring and ranking of the 11 indicators in the evaluation system with 5 dimensions, the governance effect, economic benefit, and service value of the land desertification governance modes in the study area are analyzed and evaluated, and the preferred modes are selected.

1) The ecological service values of different modes exhibit significant differences: the grape-based integrated mode stands out in economic value; the Haloxylon ammodendron-Cistanche deserticola mode demonstrates the best ecological benefits; the Lycium barbarum complex mode shows good balance, with clear coordinated development of economic and ecological benefits; the Jerusalem artichoke complex mode performs exceptionally well under specific site conditions, achieving the highest soil improvement and water-saving efficiency among all modes, although its economic output is relatively low; the Xanthoceras sorbifolium complex mode maintains balanced and stable performance across all indicators, lacking outstanding advantages but being an ideal choice for transitional zones due to its drought tolerance and woody oilseed potential; the ecological livestock farming mode delivers notable short-term benefits through a “forage-livestock-fertilizer” cycle that enables rapid returns, though carrying capacity must be strictly controlled; 2) Based on the comprehensive evaluation of ecological service value based on AHP, the order of the ecological service value scores of the 6 ecological governance modes is: Haloxylon ammodendron-Cistanche deserticola mode (comprehensive score of 88.9) > grape composite mode (85.6) > Jerusalem artichoke complex mode (82.8) > Lycium barbarum complex mode (80.5) > ecological farming mode (78.6) > X. sorbifolia complex mode (75.2).

Different governance modes exhibit significant complementarity in their ecosystem service functions: the Haloxylon ammodendron-Cistanche icola mode excels in ecological barrier function, the grape composite mode leads in economic output, and the Jerusalem artichoke composite mode demonstrates excellent performance in saline-alkali improvement and water conservation, while the remaining modes possess applicable value under specific site conditions. Based on multi-criteria comprehensive evaluation using the AHP, the Haloxylon amodendron-Cistanche deserticola mode, the grape composite mode, and the Jerusalem artichoke composite mode were selected as the optimal regional governance paradigms. Accordingly, a spatial scheme based on ecological functional zoning is proposed: developing a Haloxylon ammodendron-Cistanche deserticola ecological-economic system in the transition, deploying water-saving grape industries in the water-proximal zone, and promoting Jerusalem artichoke-Xanthoceras sorbifolia composite restoration in the salt-alkali zone.

China's pulp industry is a significant contributor to greenhouse gas emissions, and the loss of forest carbon sinks due to the process of obtaining wood raw materials is particularly significant. Domestic wood pulp production and consumption continue to rise, further intensifying carbon emissions and pressure on timber resources, posing a serious challenge to achieving industry emission reduction targets. To address this challenge, mitigate the loss of forest carbon stocks and reduce carbon emissions from the industry, this study proposes the replacement of wood pulp raw materials with giant reed raw materials in the production of pulp and evaluates the potential for emission reduction and avoidance of the loss of forest carbon stocks.

This study adopts the Life Cycle Assessment (LCA) method to quantitatively analyze the greenhouse gas (GHG) emissions of the wood pulp and giant reed pulp industry chain, and systematically assesses the potential contribution of the raw material substitution strategy (giant reed instead of wood) in the reduction of emissions in the pulp industry.

The replacement of wood by giant reed in the production of pulp significantly reduces greenhouse gas emissions. The life cycle assessment shows that the emission reduction per unit of production (1 Adt) for different pulp type (BCP, UBCP, CMP) is about 0.056-0.214 tCO₂e, which is equivalent to a reduction of GHG emissions by 3.42%-14.76%, with the raw material collection and transportation stages being the key emission reduction links. At the same time, the use of 1 ton of ruderalis pulp reduces timber harvesting by about 4.732 m³ and retains about 4.844 tCO₂e of forest carbon stocks. The analysis of wood pulp capacity planning scenarios based on 65 production lines in 46 pulp mills in China in 2025 shows that bleached chemical pulp (BCP) has the greatest potential for emission reductions, and that Shandong, Hubei, and Guangxi are key regions for emission reductions. Under the 2060 carbon neutrality goal, the full implementation of the giant reed feedstock substitution strategy is predicted to retain about 39.26-157.05 TgCO₂e of forest carbon stocks per year and significantly reduce dependence on imported wood.

Substitution of giant reed raw materials for wood raw materials for pulp production has significant synergistic benefits of greenhouse gas emission reduction and forest carbon sink protection, which not only reduces dependence on imported wood, but also provides scientific and feasible technological support for the pulp industry to implement the national 2060 carbon neutrality goal, while significantly reducing dependence on imported wood.