Mangrove ecosystems play an irreplaceable role in coastal environments by providing essential ecosystem services. Diverse mangrove species have different functions due to their morphological and physiological characteristics. A precise spatial distribution map of mangrove species is therefore crucial for biodiversity maintenance and environmental conservation of coastal ecosystems. Traditional satellite data are limited in fine-scale mangrove species classification due to low spatial resolution and less spectral information. This study employed unmanned aerial vehicle (UAV) technology to acquire high-resolution multispectral and hyperspectral mangrove forest imagery in Guangxi, China. We leveraged advanced algorithms, including RFE-RF for feature selection and machine learning models (Adaptive Boosting (AdaBoost), eXtreme Gradient Boosting (XGBoost), Random Forest (RF), and Light Gradient Boosting Machine (LightGBM)), to achieve mangrove species mapping with high classification accuracy. The study assessed the classification performance of these four machine learning models for two types of image data (UAV multispectral and hyperspectral imagery), respectively. The results demonstrated that hyperspectral imagery had superiority over multispectral data by offering enhanced noise reduction and classification performance. Hyperspectral imagery produced mangrove species classification with overall accuracy (OA) higher than 91% across the four machine learning models. LightGBM achieved the highest OA of 97.15% and kappa coefficient (Kappa) of 0.97 based on hyperspectral imagery. Dimensionality reduction and feature extraction techniques were effectively applied to the UAV data, with vegetation indices proving to be particularly valuable for species classification. The present research underscored the effectiveness of UAV hyperspectral images using machine learning models for fine-scale mangrove species classification. This approach has the potential to significantly improve ecological management and conservation strategies, providing a robust framework for monitoring and safeguarding these essential coastal habitats.

Atmospheric vapor pressure deficit (VPD, indicative of atmospheric water conditions) has been identified as a major driver of global vegetation dynamics. Drylands, including deserts, temperate grasslands, savannas, and dry forests, are more sensitive to water conditions and affect carbon, nitrogen, and water cycles. However, our knowledge is limited on the way increasing VPD affects vegetation growth and evapotranspiration (ET) in global drylands. In this study, we used long-term satellite datasets combined with multiple statistical analyses to examine the relationship between the satellite-derived normalized difference vegetation index (NDVI), a proxy for vegetation growth, and ET to VPD across global drylands. We found that significant decreases in NDVI and ET predominantly influenced the NDVI (R_{VPD − NDVI}) and ET (R_{VPD − ET}) responses to VPD in both the savannas and dry forests of South American, African, and Australian savannas and dry forests, as well as in temperate grasslands (e.g., Eurasian steppes and American prairies). Notably, more than 60% of global drylands exhibited significantly negative R_{VPD − NDVI} and R_{VPD − ET} values. In contrast, the percentage of significantly negative R_{VPD − NDVI} and R_{VPD − ET} decreased to 60° N). In predominantly warm drylands (60° N~60° S), negative VPD effects were significantly and positively regulated by soil water availability, as determined by multiple linear regression models. However, these significant regulatory effects were not observed in cold drylands. Moving-window analyses further revealed that temporal changes in R_{VPD − NDVI} and R_{VPD − ET} were positively correlated with changes in the Standardized Precipitation Evapotranspiration Index (SPEI). In warm drylands, areas with increasing R_{VPD − NDVI} and R_{VPD − ET} over time showed an increasing trend in the SPEI, whereas areas with a decreasing SPEI showed a negative trend in R_{VPD − NDVI} and R_{VPD − ET} values over time. Given the increasing atmospheric dryness due to climate change, this study highlighted the importance of re-evaluating the representation of the role of water availability in driving the response of the carbon-water cycle to increased VPD across global drylands.

Assessing vegetation phenology is very important for better understanding the impact of climate change on the ecosystem, and many vegetation index datasets from different remote sensors have been used to quantify vegetation phenology from a regional to global perspective. This study mainly analyzes the similarities and differences in phenology derived from GIMMS NDVI3g and MODIS NDVI datasets across different biomes throughout temperate China. We applied three commonly used methods to extract the start and end of the growing season (SOS and EOS) from two datasets between 2000 and 2015, and analyzed the spatio-temporal characteristics and trends of key phenological parameters between these two datasets in temperate China. Results showed that the multi-year mean GIMMS NDVI was higher than MODIS NDVI throughout most of temperate China, and the consistencies between GIMMS NDVI and MODIS NDVI for all biomes in the senescence phase were better than those in the green-up phase. NDVI differences between GIMMS and MODIS resulted in some distinctions between phenology derived from the two datasets. The results of SOS and EOS for three methods also showed wide discrepancies in spatial patterns, especially in SOS. For different biomes, differences of SOS in forests were obviously less than that in shrublands, grasslands-IM, grasslands-QT and meadows, whereas the differences of EOS in forests were relatively greater than that in SOS. Moreover, large differences of phenological trends were found between GIMMS and MODIS datasets from 2000 to 2015 in entire region and different biomes, and it is particularly noteworthy that both SOS and EOS showed a low proportion of the identical significant trends. The results suggested NDVI datasets obtained from GIMMS and MODIS sensors could induce the differences of the inversion of vegetation phenology in some degree due to the differences of instrumental characteristics between these two sensors. These findings highlighted that inter-calibrate datasets derived from different satellite sensors for some biomes (e.g., grasslands) should be needed when analyzing land surface phenology and their trends, and also provided baseline information for choosing different NDVI datasets in subsequent studies on vegetation patterns and dynamics.

Phenology and photosynthetic capacity both regulate carbon uptake by vegetation. Previous research investigating the impact of phenology on vegetation productivity has focused predominantly on the start and end of growing seasons (SOS and EOS), leaving the influence of peak phenology metrics—particularly in typical climatic transition zones—relatively unexplored. Using a 24-year (2000–2023) enhanced vegetation index (EVI) dataset from the Moderate Resolution Imaging Spectroradiometer (MODIS), we extracted and examined the spatiotemporal variation for peak of season (POS) and peak growth (defined as EVI_max) of forest vegetation in the Funiu Mountain region, China. In addition to quantifying the factors influencing the peak phenology metrics, the relationship between vegetation productivity and peak phenological metrics (POS and EVI_max) was investigated. Our findings reveal that POS and EVI_max showed advancement and increase, respectively, negatively and positively correlated with vegetation productivity. This suggested that variations in EVI_max and peak phenology both increase vegetation productivity. Our analysis also showed that EVI_max was heavily impacted by precipitation, whereas SOS had the greatest effect on POS variation. Our findings highlighted the significance of considering climate variables as well as biological rhythms when examining the global carbon cycle and phenological shifts in response to climate change.

Eucalyptus, valued for its rapid growth and economic potential, has been widely introduced in China to address timber demands while conserving natural forests. Precisely estimating the spatiotemporal expansion of short-rotation eucalyptus plantations is crucial for evaluating their ecological and social value and formulating effective sustainable forestry policies. Medium-resolution satellite images, such as Landsat data, offer a cost-effective tool for large-scale forest mapping compared to traditional forest inventories. This study used pixel-level time series analysis to identify annual eucalyptus plantation distributions across Guangxi, China, from 2004 to 2019, based on the standard temporal vegetation index curves derived from the characteristics of short-rotation and fast-growing eucalyptus. Furthermore, an image segmentation method, coupled with an empirical relationship linking patch-level landscape indices to optimal thresholds, was employed to eliminate isolated pixels and reduce omission errors arising from above time series analysis. The established thresholds increased accurate identification of eucalyptus patches within segments. Our proposed eucalyptus detection algorithm achieved an overall accuracy exceeding 80%, demonstrating its effectiveness. The analysis revealed eucalyptus plantations increased from 0.42 × 10 ⁶ ha in 200 ⁴ to 2.47 × 10 ⁶ ha in 2019, exhibiting a pronounced northward expansion. Initially concentrated in upland areas, plantations subsequently expanded into flatter terrains, raising concerns about potential agricultural conflicts. Annual eucalyptus plantation maps offer critical information for sustainable forest management and policymaking. This study highlights the potential of medium-resolution satellite data and time series analysis for robust and cost-effective monitoring of annual short-rotation timber forest expansion dynamics over large scales.

Both topography and wildfire can exert significant influences on ecosystem processes and functions during boreal forest successions. However, their impacts on ecosystem multifunctionality (EMF) remain unclear. A mega-fire burned an area of 8700 hectares in the Great Xing’an Mountains in 2000, creating a wide range of fire severity levels across various topographic positions. This provided a unique opportunity to explore the impacts of mixed-severity fire disturbance in boreal forests. We evaluated the effect pathways of wildfire and topography on aboveground multifunctionality (AEMF), soil multifunctionality (SEMF), and overall multifunctionality (OEMF). We found that high-severity burning resulted in lower AEMF, SEMF, and OEMF relative to low-severity burning. Topographic positions significantly influenced SEMF and OEMF, but not AEMF. Specifically, both lower SEMF and OEMF were observed on south-facing slopes. The structure equation model analysis showed that aspect had exerted strong indirect effects on AEMF, SEMF, and OEMF by affecting soil moisture and regenerated tree density (RTD). Fire severity had indirect negative effects on AEMF and OEMF by reducing RTD and on SEMF by reducing soil bacterial diversity and RTD. Our study elucidates the necessity of considering postfire site environments to better manage forest ecosystems and, in turn, promote the rapid recovery of boreal ecosystem functions.