A framework for integrating spatiotemporal deep learning methods with landsat for annual land cover and impervious surface mapping

Rylie Fleckenstein; Danika Fay Wellington; Suming Jin; Heather J. Tollerud; Jesslyn F. Brown; Jon Dewitz; Neal J. Pastick; Christopher P. Barber; Austin O'Brien; Mark Spanier

doi:10.1016/j.rse.2026.115347

USGS · 70274250

A framework for integrating spatiotemporal deep learning methods with landsat for annual land cover and impervious surface mapping

Rylie Fleckenstein·Danika Fay Wellington·Suming Jin·Heather J. Tollerud·Jesslyn F. Brown·Jon Dewitz·Neal J. Pastick·Christopher P. Barber·Austin O'Brien·Mark Spanier

View original source ↗Export metadata ↓

Abstract

Land cover information is essential for understanding Earth’s surface dynamics and how vegetation, water, soil, climate, and terrain interact. The National Land Cover Database (NLCD) has been the authoritative source for consistent U.S. land cover mapping. To extend NLCD’s temporal resolution and reduce production latency, we developed the Land Cover Artificial Mapping System (LCAMS)—a prototype spatiotemporal deep learning framework piloted as the foundation for the new Annual NLCD. LCAMS builds on concepts from legacy NLCD and the U.S. Geological Survey Land Change Monitoring, Assessment, and Projection (LCMAP) initiatives. It employs a loosely coupled two-stage architecture consisting of independent but functionally interdependent spatial and temporal models. Spatial models extract per-year information from Landsat data, while the temporal models refine the spatial outputs to enforce inter-annual consistency—critical for reliable land change monitoring. LCAMS produces annual 30 m resolution land cover and impervious surface outputs, with region-specific fine-tuning to generalize across diverse landscapes and temporal dynamics. Validation was conducted using an independent dataset of 1925 randomly sampled plots from five U.S. Landsat Analysis Ready Data (ARD) tiles spanning 1985-2021, selected for spatial and temporal variability. This dataset was used consistently to evaluate LCAMS, Legacy NLCD, and LCMAP. Using the NLCD legend, LCAMS achieved 72.1 ± 1.60% overall agreement, compared to 71.1 ± 1.7% agreement for Legacy NLCD. Using the LCMAP legend, LCAMS achieved 83.4 ± 1.22% agreement, compared to 84.6 ± 1.11% agreement for LCMAP. Overall, LCAMS delivers comparable accuracy while offering higher thematic resolution, longer temporal coverage, and automated production of annual 30 m CONUS land cover.

Explore related subjects

Remote Sensing of Environment Article

Keep this discovery

BibTeX RIS

Rylie Fleckenstein, Danika Fay Wellington, Suming Jin, Heather J. Tollerud, Jesslyn F. Brown, Jon Dewitz, Neal J. Pastick, Christopher P. Barber, Austin O'Brien, Mark Spanier. 2026-03-05. A framework for integrating spatiotemporal deep learning methods with landsat for annual land cover and impervious surface mapping. https://doi.org/10.1016/j.rse.2026.115347

Cite the original work for its findings. Save a collection to share your selection of sources.

Discover connections

Article subject Remote Sensing of Environment subject

Connections use source metadata and explicit phrase matches, not verified experimental comparisons.

KEEP EXPLORING

Related discoveries

USGS2026

Design and function of the Autonomous Benthic Imaging and Surveying System (ABISS) for remote sensing of lake and seabed environments

Lake and seabed environments are home to fisheries and other biota that are important to ecosystems and economies, yet these environments and the species that use them are difficult to accurately assess and monitor. Traditional benthic survey techniques, like bottom trawling used by the U.S. Geological Survey, are limited by substrate constraints, poor spatial resolution and precision, and operational depth limits, hindering accurate assessment of benthic species and habitats. In response to these limitations, the U.S. Geological Survey developed the Autonomous Benthic Imaging and Surveying System, a camera system integrated into underwater vehicles, to capture high-resolution images of the lakebed. The system uses color and stereo cameras to collect imagery, which can be analyzed using computational methods to detect organisms and (or) characterize habitat features, such as geologic substrate types. The system has been integrated into autonomous underwater vehicles and into an underwater housing used by self-contained underwater breathing apparatus (SCUBA) divers. Although the engineering of the system was motivated by the need for data collection in the Great Lakes, it has potential to collect high quality data in any aqueous setting with sufficient water clarity and safe operating conditions. The Autonomous Benthic Imaging and Surveying System can operate across diverse depths and light conditions to map and quantify ecological patterns that were difficult or impossible to assess using traditional methods. The Autonomous Benthic Imaging and Surveying System offers the potential for accurate and precise monitoring and assessment of native benthic biota, invasive species, and habitat, potentially providing natural resource managers with improved information to support decision making about benthic resource management.

Great Lakes↗

USGS2026

On-demand global Landsat evapotranspiration product: Development, evaluation, and dissemination

Global actual evapotranspiration (ET) is one of the essential climate variables needed to understand and manage the relationships among food, energy, and water resources. The U.S. Geological Survey Earth Resources Observation and Science (EROS) Center launched a provisional ET product in 2020, offering on-demand, field-scale global coverage derived from Landsat data through the EROS Science Processing Architecture (ESPA) platform. The ESPA interface provides ET data for cloud-free Landsat overpasses starting in 1982 with Landsat 4 through the current Landsat 9. The ET data are delivered as a Provisional Level-3 Science product created using the Operational Simplified Surface Energy Balance (SSEBop) model. Landsat surface temperature and reference ET are the main model drivers along with vegetation index and net radiation for model parameterization. A large volume of Landsat-based ET orders (e.g., over 1,200,000 images from June 2020 through December 2025) around the world indicate increasing awareness and application of the ET data. The ESPA platform enables land and water resource managers and researchers to access a first-order ET product without requiring advanced knowledge of remote sensing technology or evapotranspiration modeling. We present the methodology and workflow of the on-demand Landsat ET product and its performance evaluations over diverse hydro-climatic settings. The product can help estimate field-scale consumptive water use and thus quickly and consistently assess historical water use, allocation, and budget to inform water management under changing environments. Future ET data aggregated to monthly and seasonal time scales are expected to enhance integration with decision-making tools and procedures.

Remote Sensing of Environment↗

USGS2026

Regreening, restoring, and reconnecting a southwestern wetland ecosystem – the Zeedyk wetland

Alluvial wetland ecosystems are vital as biodiversity hotspots but are increasingly threatened by anthropogenic stressors and drought. These pressures are especially acute in arid and semi-arid regions, where eco-hydrologic connectivity is fragile and recovery is slow. This study quantifies the efficacy of nature-based solutions, particularly the ‘Zeedyk approach,’ which employs low-tech Natural Infrastructure in Dryland Streams (NIDS)—including rock detention structures—to slow surface water, raise groundwater tables, and restore wetland function at a spring-fed wetland in Cebolla Canyon, New Mexico, U.S.A. Our results depict a Restoration Feedback Loop that captures stages of change from a healthy wetland in 1935, altered by 20th-century agriculture and grazing, to the re-establishment of the historical flow regime by 2024 documented through an 89-year archive of aerial imagery (1935–2024). By the end of our study period, the Spring-Fed Wetland had expanded by roughly 229% of the original 1935 area, to 4.13 ha. Using 40 years of satellite data, we assess changes in vegetation and hydrology with remote sensing indices. Spatial and temporal analyses reveal significant increases in vegetation greenness and wetness, particularly in an Expanded Wetland subregion, which exhibited ∼3.5x higher wetness and ∼1.5x higher greenness trends compared to adjacent areas. Monthly metrics highlight seasonal variability, with increases in greenness linked to monsoonal rainfall and lateral water redistribution, indicating that restoration impacts extend beyond the primary wetland. This study demonstrates the utility of cloud-based platforms like Google Earth Engine and USGS EarthExplorer for long-term monitoring of wetland restoration, while quantifying the efficacy of the ‘Zeedyk approach’ and demonstrating its potential as a scalable method to restore and conserve wetland meadows in other arid and semi-arid landscapes.

New Mexico↗