Dr. Yichuan Ma's recent publication on IEEE Geoscience and Remote Sensing Magazine

Mitigating Topographic Effects in Optical Remote Sensing: From surface reflectance to high-level products 

Congratulations to Dr. Yichuan Ma, Professor Shunlin Liang, and the research team for their recent publication in the IEEE Geoscience and Remote Sensing Magazine (Early Access, 2026). The paper is available via DOI: 10.1109/MGRS.2026.3690171.

Mountainous terrains cover approximately 24% of the Earth's land surface, serving as vital hotspots for biodiversity and providing up to 60% of the world's annual freshwater. Accurate estimation of environmental variables in these regions is of immense scientific importance for global climate and hydrological monitoring. However, complex topography systematically alters the sun-target-sensor geometry through slope-aspect effects, shadowing, and terrain obstruction. Most operational satellite products, such as mainstream MODIS and Landsat pipelines, rely on flat-terrain assumptions, introducing systematic biases that can reach up to 58% in surface reflectance and greater than 1 K in land surface temperature (LST).

This review provides a comprehensive synthesis of topographic correction and topography-aware strategies based on a retrieval chain. We reviewed the progress in reducing topographic impacts in surface reflectance in three strategies: 1) empirical and semiempirical methods, 2) physically based radiative transfer correction methods, and 3) combined atmospheric–topographic correction schemes. Furthermore, we critically examine the propagation of topographic effects and specific mitigation strategies across three key downstream domains: 1) surface radiation budget components [e.g., albedo, shortwave/longwave radiation, and land surface temperature (LST)]; 2) biophysical variables [e.g., vegetation indices (VIs), leaf area index (LAI), fraction of absorbed photosynthetically active radiation (FAPAR), and gross primary production (GPP)]; and 3) geophysical variables [specifically, snow cover and evapotranspiration (ET)]. Current methodologies broadly converge into two fundamental strategies: (i) Correcting input data by removing topographic distortions at the radiometric level. It produces a "flat-equivalent" surface reflectance that can be fed into standard flat-terrain algorithms. However, it risks altering original physical signals and can propagate artifacts, such as overcorrection in shadowed slopes. (ii) Directly couples topographic effects within the retrieval algorithm itself, explicitly modifying the sun-target-sensor geometry and incorporating topography into energy balance or biophysical models. It ensures better physical consistency but is often highly complex and must be tailored to specific variables.

Despite rapid methodological advancements, the article's discussion highlights a significant gap between sophisticated research algorithms and operational execution: (i) Most operational global products (such as standard MODIS and Landsat pipelines) continue to ignore topographic effects, resulting in systematic biases in rugged regions that span ~24% of the Earth's surface. These biases inherently distort long-term climate assessments and large-scale modeling. (ii) Physics-based correction methods demand high-resolution Digital Elevation Models (DEMs) and intensive computational power, impeding their scalability for long-term time-series analyses. (iii) Validating satellite estimations in mountains is severely hindered by the scarcity of in situ data and inherent instrument biases. Furthermore, extensive subpixel heterogeneity creates a massive scale mismatch between point-based field measurements and moderate-resolution satellite pixels.

To facilitate reliable global monitoring of mountainous ecosystems, the review outlines several crucial research priorities: (i) Physics-Informed Artificial Intelligence (AI): A deep fusion of physical knowledge with advanced AI is highly advocated. (ii) Operational Integration: Moving corrections from case studies to operational productions require both high-accuracy and efficiency methods, which will enable global topographically-corrected datasets. (iii) Robust Validation Frameworks: Establishing robust validation networks will require multi-scale bridging, e.g., using high-resolution imagery to scale up ground measurements to match coarse satellite products. Leveraging 3D radiative transfer benchmark datasets will also provide consistent references to reliably evaluate algorithms in complex terrains.

Conceptual diagram illustrating how composite and solo slopes alter sun-target-sensor geometry and affect top-of-atmosphere (TOA) observations.

Global distribution of mountainous regions highlighting vast elevation gradients and diverse morphological landscapes.

Reference: Y. Ma, S. Liang, T. He, J. Wen, A. Li, H. Ma, Y. Yao, J. Xu, W. Li, and Y. Zhang, "Mitigating Topographic Effects in Optical Remote Sensing: From surface reflectance to high-level products," IEEE Geoscience and Remote Sensing Magazine, 2026, doi: 10.1109/MGRS.2026.3690171.

Keywords:

Topographic Correction, Optical Remote Sensing, Surface Energy Budget, Biophysical Variables, Complex Terrain