Search

Elevation gradients provide natural laboratories for investigating tropical tree ecophysiology in the context of climate warming. Previously observed trends with increasing elevation include decreasing stem diameter growth rates (GR), increasing leaf mass per area (LMA), higher root-to-shoot ratios (R:S), increasing leaf δ13C, and decreasing leaf δ15N. These patterns could be driven by decreases in temperature, lower soil nutrient availability, changes in species composition, or a combination thereof. We investigated whether these patterns hold within the genus Flindersia (Rutaceae) along an elevation gradient (0–1,600 m) in the Australian Wet Tropics. Flindersia species are relatively abundant and are important contributors to biomass in these forests. Next, we conducted a glasshouse experiment to better understand the effects of temperature, soil nutrient availability, and species on growth, biomass allocation, and leaf isotopic composition. In the field, GR and δ15N decreased, whereas LMA and δ13C increased with elevation, consistent with observations on other continents. Soil C:N ratio also increased and soil δ15N decreased with increasing elevation, consistent with decreasing nutrient availability. In the glasshouse, relative growth rates (RGR) of the two lowland Flindersia species responded more strongly to temperature than did those of the two upland species. Interestingly, leaf δ13C displayed an opposite relationship with temperature in the glasshouse compared with that observed in the field, indicating the importance of covarying drivers in the field. Leaf δ15N increased in nutrient-rich compared to nutrient-poor soil in the glasshouse, like the trend in the field. There was a significant interaction for δ15N between temperature and species; upland species showed a steeper increase in leaf δ15N with temperature than lowland species. This could indicate more flexibility in nitrogen acquisition in lowland compared to upland species with warming. The distinguishing feature of a mountaintop restricted Flindersia species in the glasshouse was a very high R:S ratio in nutrient-poor soil at low temperatures, conditions approximating the mountaintop environment. Our results suggest that species traits interact with temperature and nutrient availability to drive observed elevation patterns. Capturing this complexity in models will be challenging but is important for making realistic predictions of tropical tree responses to global warming.

In recent years, several catastrophic landslide events have been observed throughout the globe, threatening to lives and infrastructures. To minimize the impact of landslides, the need of landslide susceptibility map is important. The study aims to extract high-quality non-landslide samples and improve the accuracy of landslide susceptibility modelling (LSM) outcomes by applying a coupled method of ensemble learning and Machine Learning (ML). The Zigui-Badong section of the Three Gorges Reservoir area (TGRA) in China was considered in the present study. Twelve influencing factors were selected as inputs for LSM, and the relationship between each causal factor and landslide spatial development was quantitatively analyzed. A total of 179 landslides have been used in the present study. About 70% of the landslide pixels were randomly considered for training, and the remaining 30% were used for validation. Logistic Regression (LR) model was applied to produce an initial susceptibility map, and the non-landslide samples were selected within the classified low-susceptibility zone. Subsequently, two ML classifiers - the Classification and Regression Tree (CART), and the Multi-Layer Perceptron (MLP), and four coupling models - the CART-Bagging, CART-Boosting, MLP-Bagging, and MLP-Boosting, were utilized for LSM. Finally, the receiver operating characteristics (ROC) curve and statistical analysis were applied for accuracy assessment. The results show that altitude and distance to rivers were the main causal factors of landslides in the study area. The LR-MLP-Boosting performed the best with an accuracy of 0.986 followed by the LR-CARTBagging, LR-CART-Boosting, and LR-MLP-Bagging. Accuracy comparisons demonstrate that ensemble learning algorithm can notably enhance the LSM performance of ML classifiers, and the Boosting algorithm marginally outperforms the Bagging algorithm. Moreover, the LR model can effectively constrain the selection range of non-landslide samples. The non-landslide sampling method constrained by LR yields higher quality samples compared to raditional random sampling method with no constraints, which develops a more excellent LSM. [ABSTRACT FROM AUTHOR]

The development of anion-exchange membrane fuel cells (AEMFCs) has recently accelerated due to synergistic improvements yielding highly conductive membranes, stable ionomers, and enhanced alkaline electrocatalysts. However, cell durability, especially under realistic conditions, still poses a major challenge. Herein, we employ low-loadings of Pt-free Pd-based catalysts in the anode of AEMFCs and elucidate potential degradation mechanisms impacting long-term performance under conditions analogous to the real-world (high current density, H2-air (albeit CO2-free), and intermittent operation). Our high-performing AEMFCs achieve impressive performance with power densities approaching 1 W cm-2 and current densities up to 3.5 A cm-2. Over a 200 h period of continuous operation in H2-air at a current density of 600 mA cm-2, our model Pd/C-CeO2 anode cell exhibits record stability (~30 mV h-1 degradation) compared to the literature and up to 6× better stability than our Pd/C and commercial Pt/C anode cells. Following an 8 h shutdown, the Pd/C-CeO2 anode cell was restarted and continued for an additional 300 h with a higher degradation rate of ~600 mV h-1. Thorough in situ evaluations and post-stability analyses provide insights into potential degradation mechanisms to be expected during extended operation under more realistic conditions and provide mitigation strategies to enable the widespread development of highly durable AEMFCs. [ABSTRACT FROM AUTHOR]

Laser surface texturing uses a pulsed laser that is scanned on the surface, wherein each pulse creates a micro-crater through material ablation. A variety of textures can be generated depending on the laser parameters and the overlap of the laser spots. This work presents a computational model that can predict the topography of a textured surface produced using a nanosecond pulsed laser. The model involves a multi-physics approach that considers laser ablation with plasma effects and the melt pool's fluid dynamics to obtain the crater profile for a single pulse. The 3D surface profile created from the multi-physics model is mathematically superimposed to mimic the spatial overlapping of multiple pulses. The model predicts surface topography when a laser is scanned along a linear track with successive overlapping tracks. The experiments have confirmed that the proposed model has an accuracy greater than 90% in predicting surface roughness (Sa), as well as volume parameters such as core void volume (Vvc) and valley void volume (Vvv). It was observed that the variation of these surface characteristics is highly non-linear with the process parameters. Furthermore, the model is used to design engineered surfaces to modify friction coefficient, adhesion, and leakage probability. It is demonstrated that the surface parameters for functional requirements can be modified significantly just by varying the overlap of the laser spots in different directions. The proposed model can be used to create textured surfaces for various applications through an appropriate choice of laser parameters and scanning parameters. [ABSTRACT FROM AUTHOR]

The management of Kommerell's Diverticulum (KD) has been evolving from open repair to a hybrid of open and endovascular repair. While there is no consensus regarding the optimal strategy, the need for less invasive treatment with less morbidity early recovery, and improved quality of life is a common goal for both the patient and the provider. [ABSTRACT FROM AUTHOR]