Advances in GeoAI for Earth Observation and Geospatial Data in Geoscience Applications

Dear Colleagues,

The last decade has seen rapid progress in geospatial artificial intelligence (GeoAI), driven by the availability of large-scale Earth Observation (EO) archives, multi-source geospatial data, and advances in deep learning. Beyond task-specific models, the field is moving toward foundation models and self-supervised pretraining, enabling stronger transferability across regions, sensors, and applications. At the same time, multimodal learning (e.g., fusing optical, SAR, hyperspectral, LiDAR, in situ and socio-economic data) is becoming central for robust monitoring and decision support in complex geoscience settings.

The goal of this Special Issue is to collect papers (original research articles and review papers) to provide insights into AI-enabled methods and digital solutions that advance geoscience applications such as natural hazard monitoring, climate and drought analytics, land degradation and desertification assessment, carbon and ecosystem accounting, and spatiotemporal prediction. We especially welcome contributions that improve explainability, uncertainty quantification, robustness, and reproducibility, and that demonstrate generalization across sensors or geographic domains. Emerging directions in the field, such as vision–language and large language model (LLM) workflows for geospatial reasoning, generative AI for data enhancement, and physics-informed or hybrid modelling, are also encouraged.

Submissions may focus on methodological innovation, benchmark datasets and evaluation protocols, open-source tools, or compelling real-world case studies that link EO and geospatial data to actionable geoscience insights. Interdisciplinary works connecting geosciences, environmental modelling, and AI are particularly welcome.

This Special Issue will welcome manuscripts that link the following themes:

• Geospatial / EO foundation models
• Multimodal data fusion (optical–SAR–LiDAR–HSI–in situ–GIS–socio-economic)
• Spatiotemporal deep learning (e.g., ConvLSTM/Transformers) for forecasting and early warning
• Generative AI (GANs/diffusion) for super-resolution, gap-filling, domain adaptation, augmentation
• Explainable AI (XAI), uncertainty quantification, reliability, robustness, and replicability
• Physics-informed ML and hybrid modelling (data-driven + process models)
• Automated mapping: Segmentation, change detection, object detection, time-series analytics
• Hazard and risk applications (landslides, floods, fires, drought, coastal hazards)
• Land degradation, desertification, biodiversity indicators, ecosystem and carbon accounting
• AI applications in geological mapping, mineral exploration, tectonics, and subsurface modelling
• Open benchmarks, datasets, reproducible pipelines, MLOps for EO/GeoAI
• LLM/VLM-enabled geospatial reasoning, tool-use, and decision support workflows

We look forward to receiving your original research articles and reviews.

Dr. Omid Ghorbanzadeh
Prof. Dr. Pedram Ghamisi
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 250 words) can be sent to the Editorial Office for assessment.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-anonymized peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Geosciences is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 1800 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• GeoAI
• earth observation
• geospatial data
• foundation models
• multimodal fusion
• transformers
• explainable AI
• generative AI
• change detection
• susceptibility mapping