Ecological Adaptation Strategies of Desert Plants in the Farming–Pastoral Zone of Northern Tarim Basin

Sustainability, Journal Year: 2025, Volume and Issue: 17(7), P. 2899 - 2899

Published: March 25, 2025

Plant functional traits are indicative of the long-term responses and adaptations plants to their environment. However, specific mechanisms by which desert plant groups (PFGs) adjust ecological adaptation strategies cope with harsh environments remain unclear, particularly in ecologically fragile farming–pastoral zones. To address this gap, study investigates analyzes morphological chemical characteristics 13 species zone northern Tarim Basin. Through cluster analysis, these were categorized into distinct PFGs elucidate response at a higher organizational level. The results as follows: (1) Based on traits, classified acquisitive, medium, conservative PFGs. These exhibited significant differences element content proportion, well adjustments (p < 0.05). (2) acquisitive group maintained high resource acquisition turnover through leaf area phosphorus content; medium occupied limited resources greater height canopy width, whereas low growth rates but investment ensure survival. Moreover, led selection divergent central different (3) Low soil nutrient availability salinization, rather than groundwater depth, identified primary environmental factors driving differentiation zone. findings suggest that arid regions employ diverse pressures. This research provides valuable insights recommendations for conservation restoration communities.

Language: Английский

---

The utility of dynamic forest structure from GEDI lidar fusion in tropical mammal species distribution models

Frontiers in Remote Sensing, Journal Year: 2025, Volume and Issue: 6

Published: May 12, 2025

Remote sensing is an important tool for monitoring species habitat spatially and temporally. Species distribution models (SDM) often rely on remotely-sensed geospatial datasets to predict probability of occurrence infer preferences. Lidar measurements from the Global Ecosystem Dynamics Investigation (GEDI) are shedding light three dimensional forest structure in regions world where this aspect has previously been poorly quantified. Here we combine a large camera trap dataset mammal Borneo Sumatra with diverse set data 47 species. Multi-temporal GEDI predictors were created through fusion Landsat time series, extending back year 2001. The availability these GEDI-based other temporally-resolved predictor variables enabled temporal matching occurrences hindcast predictions at years 2001 2021. Our GEDI-Landsat approach worked well metrics related canopy height (relative 95th percentile returned energy R 2 = 0.62 relative RMSE 41%) but, not surprisingly, was less accurate interior vegetation (e.g., plant area volume density 0 5 m above ground 0.05 85%). For SDM analyses, tested several combinations sets found that when considering pool multiscale predictors, exact composition, whether Fusion included, didn’t have impact generalized linear modeling (GLM) Random Forest (RF) model performance. Adding baseline only meaningfully improved performance some (n 4 RF n 3 GLM). However, used smaller more suitable hindcasting occurrence, SDMs showed meaningful improvements 9 GLM) importance increased they combined set. Moreover, as examined partial dependence, utility evident regards ecological interpretability. We produced catalog maps all mammals 90 spatial resolution 2021, enabling subsequent interpretation conservation analyses.

Language: Английский

---

Choosing the Optimal Global Digital Elevation Model for Stream Network Delineation: Beyond Vertical Accuracy

Earth and Space Science, Journal Year: 2024, Volume and Issue: 11(12)

Published: Nov. 27, 2024

Abstract Satellite‐derived global digital elevation models (DEMs) are essential for providing the topographic information needed in a wide range of hydrological applications. However, their use is limited by spatial resolution and vertical bias due to sensor limitations observing bare terrain. Significant efforts have been made improve DEMs (e.g., TanDEM‐X) create bare‐earth FABDEM, MERIT, CEDTM). We evaluated accuracy Central European mountains submontane regions, assessed how DEM resolution, vegetation offset removal, land cover, terrain slope affect stream network delineation. Using lidar‐derived DTM national networks as references, we found that: (a) outperform across all cover types. RMSEs increased with increasing non‐forest areas. In forests, however, negative effect was outweighed even DTMs; (b) derived affected more than DEMs. Stream delineation performed poorly areas relatively well forests. Increasing improved streams performance; (c) using higher 12 m delineation, but also need effective removal. Our results indicate that alone does not reflect perform This underscores include performance quality rankings.

Language: Английский

---

Naturalness indicators of forests in Southern Sweden derived from the canopy height model

European Journal of Remote Sensing, Journal Year: 2024, Volume and Issue: 58(1)

Published: Dec. 23, 2024

Forest canopies embody a dynamic set of ecological factors, acting as pivotal interface between the Earth and its atmosphere. They are not only result an ecosystem's ability to maintain inherent processes, structures, functions but also reflection human disturbance. This study introduces methodology for extracting comprehensive human-interpretable features from Canopy Height Model (CHM) with resolution 1 meter. These then analyzed identify reliable indicators degree naturalness forests in Southern Sweden. Using these features, machine learning models – specifically, perceptron, logistic regression, decision trees trained examples exhibiting known high low degrees naturalness. achieve prediction accuracies ranging 89% 95% on unseen data, depending area region interest. The predictions proposed method easy interpret, making them particularly valuable various stakeholders involved forest management conservation.

Language: Английский

---