IEEE J-STARS Announces Special Issue on AI Applications in Wetlands Remote Sensing

By Kevin P. Corbley

In December 2025, the IEEE Journal of Selected Topics in Applied Earth Observation and Remote Sensing (J-STARS) announced a special issue on “Artificial Intelligence in Coastal Wetlands Remote Sensing” to showcase how AI and deep learning techniques can enhance geospatial technologies to map, monitor, and model fragile wetland ecosystems.

IEEE J-STARS has chosen five guest editors for the issue, which is accepting submissions now through May 31, 2026. The topic is timely because coastal wetlands – salt marshes, mangroves, tidal flats, and estuarine systems – are as fragile and dynamic as they are economically and ecologically valuable. They are also under unprecedented stress.

As stated in the Call for Papers, “[T]hese fragile ecosystems are increasingly threatened by sea-level rise, land use change, pollution, and other anthropogenic pressures…Recent advances in AI, deep learning, and data science have opened up transformative opportunities to harness the vast potential of multi-sensor, high-resolution, and time-series remote sensing data for wetland research and conservation.”

Why Wetlands, Why Now?

“Wetlands are very important; we often call wetlands the kidneys of the Earth,” said Dr. Huiying Li, Associate Professor at China’s Qingdao University of Technology. Her research within the School of Environmental and Municipal Engineering focuses on monitoring invasive species in coastal wetlands. She is a guest editor for the special issue.

Dr. Li explained the “Kidney” metaphor is accurate because wetlands are biologically productive and regulate natural systems on Earth in ways that sustain ecological and human well-being. They store and slowly release water, buffering communities from floods during wet seasons and maintaining water supply during dry periods. They trap sediments, filter pollutants, and remove excess nutrients, functioning as natural purification systems that protect downstream water quality.

“Wetlands also play an important role in climate regulation,” she added. “They store large amounts of carbon in their soils and play an important role in the carbon cycle.”

They also provide essential habitat for fish, migratory birds, and countless other species that depend on the unique hydrological and ecological conditions wetlands create. Beyond ecological value, wetlands economically support fisheries, tourism, and cultural traditions, offering economic and social benefits to communities worldwide.

The attractive environment created by wetlands, especially along coastlines, appeals to humans, and this is a major source of ecosystem stress, explained Dr. Nishan Bhattarai, Assistant Professor in the University of Oklahoma’s Department of Geography and Environmental Sustainability. Also a J-STARS guest editor, his research focuses on remote sensing and hydrology.

“A lot of those pressures that wetlands face relate to human activities, such as urbanization, sea-level rise, climate change…and even agricultural expansion as the demand for food increases,” Dr. Bhattarai said.

Urban development converts coastal and inland wetlands into built environments, replacing natural water‑regulating systems with impermeable surfaces, he explained. Agricultural expansion drains wetlands to create farmland, often degrading soil health, reducing long‑term productivity, and culminating in pesticides, fertilizers, and animal waste draining into the wetland systems.

“Anything that changes the natural flow of water is going to affect the wetland,” he said.

Why AI and Deep Learning

Wetlands have been mapped with remotely sensed data for as long as there have been EO satellites in orbit. But truly analyzing and understanding the incredibly complex interaction of natural processes and external forces at work across the ecosystem have been elusive. AI and deep learning technologies are game changers because of their ability to identify, correlate, and track multiple variables extracted from numerous data sets.

“Coastal wetlands monitoring today is moving toward interpreting air, space, and ground observations at large scale…at higher resolution and revisit frequency rates using AI and deep learning,” said Dr. Li.

No single geospatial data set provides all the answers. Optical and radar satellite data are crucial, as are airborne LiDAR elevation points and terrestrial data collection, including measurements of water levels and soil moisture. These data sets become more valuable – and difficult to correlate – when captured consistently over time. The challenge is not the lack of data, but the complexity of integrating these diverse sources into a coherent, actionable picture.

AI and deep learning models excel at processing massive, multi‑modal datasets to extract patterns that would be impossible to detect manually. Deep learning algorithms can detect subtle changes in vegetation or hydrology with unprecedented accuracy. Time‑series analysis powered by AI allows scientists to track seasonal cycles, long‑term trends, and abrupt disturbances.

The real impact of AI, Dr. Bhattarai concluded, is that instead of reacting only to wetland degradation after it occurs, AI enables proactive modeling. Predictive systems based on remote sensing data can forecast how wetlands will respond to storms, heavy rainfall, sea‑level rise, or land‑use changes.

Professors Li and Bhattarai agree that AI systems can predict response to events and also identify areas at highest risk—whether from pollution, encroaching development, hydrological disruption, or climate‑driven impacts. These insights support early intervention, targeted restoration, and more resilient planning for coastal environments.

Call for Submissions

The J-STARS special issue will cover broad topics related to wetland monitoring and conservation through the integration of EO and AI – with emphasis on the shift from reactive to predictive management. The Call for Papers seeks submissions on these and other AI-related wetland topics:

• Deep learning techniques for coastal wetland classification and change detection
• Multi-source data fusion (e.g., SAR, optical, LiDAR) for ecosystem monitoring
• Time-series analysis and spatiotemporal modeling of wetland dynamics
• AI-supported quantification of blue carbon stocks and greenhouse gas fluxes
• AI applications in modeling hydrological, geomorphological, and biogeochemical processes High-resolution and large-scale coastal wetland observation using AI
• Early warning systems for wetland vulnerability and resilience
• AI-enhanced assessments of wetland degradation, restoration, and connectivity
• Biodiversity monitoring and invasive species detection through intelligent sensing