RESUMEN
Alpine wetlands are critical ecosystems for global carbon (C) cycling and climate change mitigation. Ecological restoration projects for alpine grazing wetlands are urgently needed, especially due to their critical role as carbon (C) sinks. However, the fate of the C pool in alpine wetlands after restoration from grazing remains unclear. In this study, soil samples from both grazed and restored wetlands in Zoige (near Hongyuan County, Sichuan Province, China) were collected to analyze soil organic carbon (SOC) fractions, arbuscular mycorrhizal fungi (AMF), soil properties, and plant biomass. Moreover, the Tea Bag Index (TBI) was applied to assess the initial decomposition rate (k) and stabilization factor (S), providing a novel perspective on SOC dynamics. The results of this research revealed that the mineral-associated organic carbon (MAOC) was 1.40 times higher in restored sites compared to grazed sites, although no significant difference in particulate organic carbon (POC) was detected between the two site types. Furthermore, the increased MAOC after restoration exhibited a significant positive correlation with various parameters including S, C and N content, aboveground biomass, WSOC, AMF diversity, and NH4+. This indicates that restoration significantly increases plant primary production, litter turnover, soil characteristics, and AMF diversity, thereby enhancing the C stabilization capacity of alpine wetland soils.
RESUMEN
Weeping cypress is an endemic tree species that is widely planted in China, and the simple stand structure and fragile ecosystem of its plantation are common issues. Exploring the effect of different gap sizes on the soil bacterial community structure of weeping cypress plantations can provide a theoretical basis for the near-natural management of forest plantations. We, therefore, constructed three kinds of forest gaps with different sizes in weeping cypress plantations, namely, small (50-100 m2), medium (100-200 m2), and large gaps (400-667 m2), for identifying the key factors that affect soil bacterial communities following the construction of forest gaps. The results suggested that the herb layer was more sensitive than the shrub layer, while the Simpson, Shannon, and richness indices of the herb layer in plots with gaps were significantly higher than those of designated sampling plots without any gaps (CK). The presence of large gaps significantly increased the understory plant diversity and the Shannon and Simpson indices of the soil bacterial alpha diversity. There were obvious changes in the community composition of soil bacteria following the construction of forest gaps. The dominant bacterial phyla, orders, and functions were similar across the plots with different gap sizes. Of the indicator bacterial species, the abundance of the nitrogen-fixing bacteria, Lysobacter_ yangpyeongensis, and Ensifer_meliloti, was significantly different across plots with different gap sizes and accounted for a large proportion of the bacterial population of plots with medium and large gaps. The understory plant diversity was mostly related to the soil bacterial community than to other soil factors. The results of structural equation modeling indicated that the understory plant diversity was the most important environmental factor in driving the composition and diversity of bacterial communities. The construction of forest gaps significantly improved the understory plant diversity, physicochemical properties of the soil, and bacterial diversity in weeping cypress plantations, and the results of the comprehensive evaluation were in the order: large gaps > small gaps > medium gaps > CK. Our results suggested that large gaps are beneficial for the diversity of above-ground plant communities and underground soil bacterial communities.
RESUMEN
Thinning can significantly promote forest productivity and ecological function. Rhizosphere fungi play an indispensable role in regulating nutrient cycling between plants and the environment, and their community composition can positively respond to anthropogenic disturbance. However, the initial effects of thinning on rhizosphere fungal community assembly have seldom been reported. In this research, we studied the alterations in the rhizosphere fungal communities of 29-year-old Pinus massoniana in East Sichuan 2 years after three different thinning intensity treatments. In addition, the responses of fungal community and functional group composition to alterations in understory vegetation and soil physiochemical properties were analyzed. Three thinning intensities were set, which were 0 (CK), 25% (LIT), and 50% (HIT), respectively. The results suggested that the richness index and Shannon index of understory vegetation increased significantly with increasing thinning intensity. The alpha diversity indices of rhizosphere fungal community and soil physiochemical properties did not show significant differences among the three treatments. The relative abundances of 17 fungal indicator species varied regularly with increasing thinning intensity, and most of them belong to Hypocreales and Eurotiales, indicating that these two orders were potential indicators for different thinning treatments. Rhizosphere fungal community assembly was determined by deterministic process, and it was driven by the diversity of understory vegetation in the initial stage of thinning. The Simpson index and Pielou index of herbs were useful measures of the main environmental factors driving the differentiation of fungal functional group composition. Based on network analysis, thinning resulted in distinct co-occurrence patterns of rhizosphere fungal functional groups. This research elucidates the initial role of thinning in rhizosphere fungal community assembly of P. massoniana and has practical significance for the functional restoration and protection of local forest ecosystem.
RESUMEN
Species composition and diversity of undergrowth vegetation community under different thinning intensities (0, 10%, 20%, 30%, 40%, 50%) were examined at the initial stage of thinning in 29-year-old Pinus massoniana plantation in the low mountain region of eastern Sichuan. The results show that all the thinning treatments could reduce the absolute dominance of Miscanthus sinensis and Dicranopteris dichotoma. The dominant species composition of shrubs in each treatment was different. There were more extensive species in the medium thinning intensity (20%, 30% and 40%) treatments than other treatments. The diversity indices increased first and then decreased with increasing thinning intensity. The variation degree of herbs was stronger than shrubs. The diversity indices of herbs were positively correlated with soil water content. The explanation amount of thinning intensity and soil physicochemical properties to community differentiation was 81%. The vegetation communities in the medium thinning intensity forests showed positive correlation with all the factors except total phosphorus. At the initial stage of thinning, herbaceous communities were more sensitive to disturbance than shrub communities. The 40% thinning intensity treatment was more closely related to soil environmental factors, with high stability and the most abundant species, which would be the best thinning measure under the experimental condition.
Asunto(s)
Pinus , Biodiversidad , China , Bosques , Fósforo , SueloRESUMEN
At present, many Deep Neural Network (DNN) methods have been widely used for hyperspectral image classification. Promising classification results have been obtained by utilizing such models. However, due to the complexity and depth of the model, increasing the number of model parameters may lead to an overfitting of the model, especially when training data are insufficient. As the performance of the model mainly depends on sufficient data and a large network with reasonably optimized hyperparameters, using DNNs for classification requires better hardware conditions and sufficient training time. This paper proposes a feature fusion and multi-layered gradient boosting decision tree model (FF-DT) for hyperspectral image classification. First, we fuse extended morphology profiles (EMPs), linear multi-scale spatial characteristics, and nonlinear multi-scale spatial characteristics as final features to extract both special and spectral features. Furthermore, a multi-layered gradient boosting decision tree model is constructed for classification. We conduct experiments based on three datasets, which in this paper are referred to as the Pavia University, Indiana Pines, and Salinas datasets. It is shown that the proposed FF-DT achieves better performance in classification accuracy, training conditions, and time consumption than other current classical hyperspectral image classification methods.
