Dodangeh et al. (2025) Analyzing climatic anomalies and ecological impacts on wetlands environmental conditions using LSTM and remote sensing imagery

Identification

• Journal: Advances in Space Research
• Year: 2025
• Date: 2025-12-12
• Authors: Parisa Dodangeh, Reza Shah–Hosseini, Saeid Homayouni
• DOI: 10.1016/j.asr.2025.12.034

Research Groups

• School of Surveying and Geospatial Engineering, College of Engineering, University of Tehran, Tehran, Iran
• Centre Eau Terre Environnement, Institut National de la Recherche Scientifique, Quebec G1K 9A9, Canada

Short Summary

This study employs deep learning models and remote sensing to detect climatic anomalies in the Anzali Wetland (Iran) from 2019-2023 and assess their ecological impacts. The findings reveal significant environmental degradation, including reduced water availability, poor vegetation health, and increased emissions, linked to these anomalies.

Objective

• To detect climatic anomalies in the Anzali Wetland using deep learning and remote sensing, and to analyze their subsequent ecological impacts on the wetland's environmental conditions.

Study Configuration

• Spatial Scale: Anzali Wetland, northern Iran.
• Temporal Scale: 2019 to 2023.

Methodology and Data

• Models used: Autoencoder-LSTM, CNN-LSTM, BiLSTM (CNN-LSTM outperformed).
• Data sources: Ground-based meteorological data, satellite remote sensing observations (Chlorophyll-a (Chl-a), Normalized Difference Vegetation Index (NDVI), Land Surface Water Index (LSWI), groundwater levels, methane (CH4) emissions, nitrogen dioxide (NO2) emissions).

Main Results

• The CNN-LSTM model demonstrated superior performance in anomaly detection, achieving the lowest Mean Squared Error (MSE) and highest R2 compared to Autoencoder-LSTM and BiLSTM.
• Climatic anomalies, primarily occurring in winter and transitional seasons, were associated with intense weather systems such as cyclones and cold fronts.
• Ecological impacts included reduced water availability and poor vegetation health, indicated by lower NDVI and NDPI values, alongside declining groundwater levels.
• A significant shift from wetland to drier land types was observed after 2021.
• Spikes in methane (CH4) and nitrogen dioxide (NO2) emissions highlighted the combined impacts of climate variability and anthropogenic activity.

Contributions

• Proposes an effective deep learning framework, specifically highlighting the CNN-LSTM model, for accurate climatic anomaly detection in wetland ecosystems.
• Provides valuable insights into the strong link between climatic anomalies and ecosystem stability in the Anzali Wetland.
• Emphasizes the critical need for integrated monitoring strategies to mitigate the negative impacts of climate change on wetlands.
• Offers a framework that supports more effective wetland conservation and sustainable land and water management practices.

Funding

• Not specified in the provided text.

Citation

@article{Dodangeh2025Analyzing,
  author = {Dodangeh, Parisa and Shah–Hosseini, Reza and Homayouni, Saeid},
  title = {Analyzing climatic anomalies and ecological impacts on wetlands environmental conditions using LSTM and remote sensing imagery},
  journal = {Advances in Space Research},
  year = {2025},
  doi = {10.1016/j.asr.2025.12.034},
  url = {https://doi.org/10.1016/j.asr.2025.12.034}
}