(Press-News.org) Accurate classification of wetland vegetation is essential for biodiversity conservation and carbon cycle monitoring. This study developed an adaptive ensemble learning (AEL-Stacking) framework that combines hyperspectral and light detection and ranging (LiDAR) data captured by UAVs to precisely identify vegetation species in karst wetlands. The approach achieved up to 92.77% accuracy—substantially outperforming traditional models—and revealed how spectral and structural features jointly improve ecosystem mapping and restoration strategies.
Karst wetlands are globally significant ecosystems that regulate water, store carbon, and harbor rich biodiversity. However, the intricate vegetation composition and similar canopy spectra among species hinder accurate remote sensing classification. Traditional field surveys are costly and spatially limited, while multispectral imaging lacks sufficient spectral resolution for species-level mapping. Light detection and ranging (LiDAR) provides 3D structural data but struggles with water-surface reflectance and weak signals. Due to these challenges, integrating complementary optical and structural data is necessary to conduct in-depth research on precise vegetation species classification in karst wetlands.
Researchers from the Guilin University of Technology and collaborators published (DOI: 10.34133/remotesensing.0452) their findings in Journal of Remote Sensing on October 16, 2025. The study introduces a UAV-based approach that merges hyperspectral imagery (HSI) and LiDAR point-cloud data through an adaptive ensemble learning stacking (AEL-Stacking) model. This innovative framework not only enhances classification accuracy but also uses local interpretable model-agnostic explanations (LIME) to visualize how each feature contributes to the decision-making process—offering both high precision and interpretability in mapping complex wetland vegetation structures.
The study demonstrated that combining HSI and LiDAR data achieved the highest overall accuracy (87.91%–92.77%), surpassing single-data approaches by up to 9.5%. The AEL-Stacking model, integrating Random Forest, LightGBM, and CatBoost classifiers, outperformed both conventional ensemble and deep-learning (Swin Transformer) algorithms by 0.96%–7.58%. The LiDAR features—especially digital surface model (DSM) variables—were pivotal for distinguishing species with distinct vertical structures, while hyperspectral vegetation indices such as NDVI and blue-edge parameters enhanced recognition of herbaceous species. These results highlight the synergy between optical and structural data in resolving species with overlapping spectral signatures.
Field surveys were conducted in the Huixian Karst Wetland of Guilin, China, one of the country's largest karst wetlands. UAV flights equipped with Headwall Nano-Hyperspec and DJI Zenmuse L1 LiDAR sensors collected over 4,500 hyperspectral images and dense point clouds (208 points/m²). The integrated dataset covered 13 vegetation types, including lotus, miscanthus, and camphor trees. Through recursive feature elimination and correlation analysis, 40 optimal features were selected from more than 600 variables. The AEL-Stacking framework adaptively tuned hyperparameters, selected the best-performing base learner as the meta-model, and validated results using 10-fold cross-validation. LIME analysis revealed DSM and blue spectral bands as the most influential features, with Lotus and Miscanthus achieving classification F1-scores above 0.9. The model significantly reduced misclassification between morphologically similar species, offering detailed vegetation maps critical for ecosystem monitoring.
"Our approach bridges the gap between spectral and structural sensing," said Dr. Bolin Fu, corresponding author. "By combining UAV hyperspectral and LiDAR data through adaptive ensemble learning, we achieved both precision and interpretability in vegetation mapping. The framework not only improves species recognition in complex karst environments but also provides a generalizable tool for ecological monitoring and habitat restoration worldwide".
The team developed the AEL-Stacking model by combining Random Forest, LightGBM, and CatBoost classifiers under a grid-search-optimized adaptive framework. The model used 70% of data for training and 30% for testing, supported by 10-fold cross-validation. Hyperspectral features (e.g., NDVI, EVI, CIg) and LiDAR-derived metrics (e.g., DSM, intensity skewness) were fused into a multidimensional dataset. To interpret results, the LIME algorithm quantified the contribution of each feature and visualized how data variations influenced species classification across multiple vegetation types.
This integrative framework demonstrates a scalable and explainable approach for high-resolution wetland mapping, potentially applicable to forest, grassland, and coastal ecosystems. Future work will focus on integrating multi-temporal UAV observations and satellite data fusion to monitor seasonal vegetation dynamics and climate-driven changes in wetland health. By enhancing the transparency and accuracy of AI-driven ecological models, this research paves the way for smarter environmental management and supports the global agenda for biodiversity conservation and carbon neutrality.
###
References
DOI
10.34133/remotesensing.0452
Original Source URL
https://spj.science.org/doi/10.34133/remotesensing.0452
Funding information
This study was supported by the National Natural Science Foundation of China (grant number 42371341), the Natural Science Foundation of Guangxi Zhuang Autonomous Region (grant number 2024GXNSFAA010351), the Innovation Project of Guangxi Graduate Education (grant number YCBZ2024179), and the Key Laboratory of Tropical Marine Ecosystem and Bioresource, Ministry of Natural Resources (grant number 2023ZD02).
About Journal of Remote Sensing
The Journal of Remote Sensing, an online-only Open Access journal published in association with AIR-CAS, promotes the theory, science, and technology of remote sensing, as well as interdisciplinary research within earth and information science.
END
Integrating light and structure: Smarter mapping for fragile wetland ecosystems
2025-12-22
ELSE PRESS RELEASES FROM THIS DATE:
ACA-SIM: A robust way to decode satellite signals over complex waters
2025-12-22
A new study introduces ACA-SIM (atmospheric correction based on satellite–in situ matchup data), a neural-network-based atmospheric correction algorithm that uses real satellite–Aerosol Robotic Network-Ocean Color (AERONET-OC) matchups to improve the accuracy of atmospheric correction over coastal waters. By learning from real-world satellite radiance and in-situ reflectance data, ACA-SIM significantly reduces errors and striping artifacts in ocean color products, outperforming existing models in turbid water and complex-aerosol conditions such as the Bohai Sea, North Africa dust ...
Probiotics can restore gut microbiome in breastfed infants
2025-12-22
Washington, D.C.—In recent years, scientists have learned that key beneficial infant gut bacteria Bifidobacterium infantis are disappearing from infants in high-resource areas such as the United States and Europe. Now, a new study published in the journal mSphere found that supplementing exclusively breastfed infants with a probiotic, B. infantis EVC001, between 2 and 4 months of age can successfully restore beneficial bacteria in their gut.
“The REMEDI study shows that it’s not too late to restore a healthy gut microbiome in breastfed infants. B. infantis can successfully take hold even after the newborn period,” ...
AI could help predict nutrition risks in ICU patients, study finds
2025-12-22
New York, NY [December 22, 2025]—A new study by researchers at the Icahn School of Medicine at Mount Sinai suggests that artificial intelligence (AI) could help predict which critically ill patients on ventilators are at risk of underfeeding, potentially enabling clinicians to adjust nutrition early and improve patient care. Details of the study were published in the December 17 online issue of Nature Communications.
The first week on a ventilator is especially important for providing proper nutrition, since patients’ needs often shift ...
Federal EITC has unexpected result, researchers say – it decreases domestic violence
2025-12-22
Fifty years since the federal earned income tax credit went into effect and a team of researchers from UConn and City University of New York have identified an unintended benefit of the antipoverty program – a significant reduction in rates of intimate partner violence among unmarried women.
“There’s been enough literature to show that good things happen because of the earned income tax credit, but what was surprising to us is the relatively large effect it has on intimate partner violence,” says UConn’s David Simon, an associate professor of economics and study co-author. ...
Researchers identify gene that calms the mind and improves attention in mice
2025-12-22
Attention disorders such as ADHD involve a breakdown in our ability to separate signal from noise. The brain is constantly bombarded with information, and focus depends on its ability to filter out distractions and detect what matters. Stimulant medications improve attention by boosting activity in circuits known to govern attention, such as the prefrontal cortex. But a new study reveals a surprising alternative: reduce background activity as a way of turning down extraneous noise.
In a paper published in Nature Neuroscience, ...
Artificial metabolism turns waste CO2 into useful chemicals
2025-12-22
In a breakthrough that defies nature, Northwestern University and Stanford University synthetic biologists have created a new artificial metabolism that transforms waste carbon dioxide (CO2) into useful biological building blocks.
In the new study, the team engineered a biological system that can convert formate — a simple liquid molecule easily made from CO2 — into acetyl-CoA, a universal metabolite used by all living cells. As a proof of concept, the engineers then used the same system to convert acetyl-CoA into malate, ...
Ancient sea anemone sheds light on animal cell type evolution
2025-12-22
One of the biggest quests in biology is understanding how every cell in an animal’s body carries an identical genome yet still gives rise to a kaleidoscope of different cell types and tissues. A neuron doesn’t look nor behave like a muscle cell but has the same DNA.
Researchers think it comes down to how cells allow different parts of the genome to be read. Controlling these permissions are regulatory elements, regions of the genome which switch genes on or off. A detailed overview of how they do this is largely restricted to a handful of classic model organisms like mice and fruit flies.
For the first time, researchers have created ...
Begging gene leads to drone food
2025-12-22
Is complex social behaviour genetically determined? Yes, as a team of biologists from Heinrich Heine University Düsseldorf (HHU) together with colleagues from Bochum and Paris discovered while studying bees. They identified a genetic factor that determines the begging behaviour of drones, which they use to obtain food socially. They are now publishing their findings in the scientific journal Nature Communications.
Male bees, or “drones”, do not have an easy time obtaining essential proteins. This ...
How climate policies that incentivize and penalize can drive the clean energy transition
2025-12-22
A new study from a team of researchers that includes faculty from the University of California San Diego and Princeton University shows how a mix of subsidies for clean energy and taxes on pollution can significantly reduce greenhouse gas emissions that cause climate change.
While these kinds of policy mixes are widely used in the real world, the the study, published in Nature Climate Change is the first to show how the combination of such policies can be simulated in economic models that are the backbone of nearly all climate policy discussions – including ...
Can community awareness campaigns in low-resource areas improve early diagnosis of colorectal cancer?
2025-12-22
In low-resource regions such as Nigeria, most people with colorectal cancer are diagnosed too late for curative treatment options. A community awareness campaign in the country helped clinicians detect both early-stage and advanced colorectal cancer in patients who had not been aware that they should be screened. That’s according to a study published by Wiley online in CANCER, a peer-reviewed journal of the American Cancer Society.
For the study, investigators ...