Grazing intensity (GI) is an important indicator for grazing situations in pastoral areas. However, it has been difficult to be observed directly in the field, due to the randomness and dynamics of the grazing behavior of livestock. Consequently, the lack of actual GI information has become a common issue in studies on quantitatively estimating GI. In this paper, a novel quantitative estimation method is proposed based on the Space-Air-Ground integrated monitoring technology. It systematically integrates GPS tracking technology, Unmanned Aerial Vehicle (UAV) observation technology, and satellite remote sensing technology. Taking Xiangdong Village on the Zoige Plateau as a study area, the trajectory data and UAV images were acquired by the GPS tracking experiments and UAV observation experiments, respectively. The GI at paddock scale (PGI) was then generated with the Kernel Density Estimation (KDE) algorithm and the above data. Taking the generated PGI as training data, an estimation model of GI at region scale (RGI) was constructed by using the time-series satellite remote sensing images and random forest regression algorithm. Finally, the time-series RGI data with a spatial resolution of 10 m in Xiangdong Village were produced by the above model. The accuracy assessment demonstrated that the generated time-series RGI data could reflect the spatial-temporal heterogeneity of actual GI, with a mean absolute error of 0.9301 and r² of 0. 8573. The proposed method provides a new idea for generating the actual GI on the ground and the time-series RGI data. This study also highlights the feasibility and potential of using the Space-Air-Ground integrated monitoring technology to generate time-series RGI data with high spatial resolution. The generated time-series RGI data would provide data support for the formulation of policies and plans related to the sustainable development of animal husbandry.

Canopy reflectance models over sloping terrain are critical and reliable tools for vegetation biophysical parameter retrieval. However, the applicability of existing models is strictly limited by individual canopy structure, such as continuous or discontinuous canopies, and hybrid vegetation structure is seldom considered in current models. Therefore, this defect greatly limits the application of canopy reflectance models in rugged terrains. To overcome this problem, we developed a canopy reflectance model suitable for both continuous and discontinuous canopies over sloping terrains (BOST). The scattering contributions of different canopies can be estimated by incorporating clumping indices; thus, we designed an algorithm to convert the clumping indices from flat terrain to sloping terrains, which explicitly considers the terrain effect on radiative transfer within canopies. In model validations, multiple schemes were used to obtain the objective evaluation results, including analyzing the terrain induced clumping index variations, comparing with the computer simulation model, typical canopy reflectance models, and the image reflectance in a real scene. The results suggested that the BOST model can capture the pattern of terrain-induced scattering components and reflectance distortions, and be successfully used in simulating the reflectance in a real mountainous region [RMSE (R ² ) values of 0.0016 (0.889), 0.0183 (0.932), and 0.0023 (0.903) in the red, NIR, and green bands, respectively]. BOST performed better in terms of accuracy and efficiency than the typical models. Further testing of BOST in topographic normalization and biophysical parameter retrieval showed its practical usage potential. These validations confirmed the good performance of BOST in both continuous and discontinuous vegetation scenes and indicated it can provide a promising tool to retrieve biophysical parameters in mountainous areas.

Vegetation biophysical products offer unique opportunities to examine long-term vegetation dynamics and land surface phenology (LSP). It is important to understand the time-series performances of various global biophysical products for global change research. However, few endeavors have been dedicated to assessing the performances of long-term change characteristics or LSP extraction derived from different satellite products, especially in mountainous areas with highly fragmented and rugged surfaces. In this paper, we assessed the time-series characteristics and LSP detections of Global LAnd Surface Satellite (GLASS) leaf area index (LAI), fractional vegetation cover (FVC), and gross primary production (GPP) products across the Three-River Source Region (TRSR). The performances of products’ temporal agreements and their statistical relationship as a function of topographic indices and heterogeneous pixels, respectively, were investigated through intercomparison among three products during the period 2000 to 2018. The results show that the phenological differences between FVC and two other products are beyond 10 days over more than 35% of the pixels in TRSR. The long-term trend of FVC diverges significantly from GPP and LAI for 13.96% of the total pixels, and the percentages of mismatched pixels between FVC and two other products are 33.24% in the correlation comparison. Moreover, good agreements are observed between GPP and LAI, both in terms of LSP and interannual variations. Finally, the LSP and long-term dynamics of the three products exhibit poor performances on heterogeneous surfaces and complex topographic areas, which reflects the potential impacts of environmental factors and algorithmic imperfections on the quality and performances of different products. Our study highlights the spatiotemporal disparities in detections of surface vegetation activity in mountainous areas by using different biophysical products. Future global change studies may require multiple high-quality satellite products with long-term stability as data support.

Topography affects the fraction of scene components of the canopy and background, resulting in the observed reflectance distortion. Modeling the canopy reflectance over rugged terrain needs to account for topographic effects. For this purpose, the existing models greatly increased the mathematical complexity while improving description of terrain and crown structure, which dramatically decreased the computational efficiency so as to limit their universal application. In this study, we developed a simplified geometric-optical model (SGOT) for simulating the scene components over rugged terrain. The geotropism of tree growth was considered to make SGOT physically sound. The internal structure of crown was simplified to make SGOT mathematically simpler. Scene component observations derived from Persistence of Vision Ray-tracer (POV-Ray) on surfaces with different normal directions and simulations were made using Geometric-Optical and Mutual Shadowing Coupled with Topography Model (GOMST) and Geometric-Optical for Sloping Terrains Model GOST; models were combined to test the SGOT model. In addition, topographic factors and crown density effect on the scene components modeling were analyzed. The results indicated that SGOT has good accuracy (R² for the areal proportions of sunlit crown (Kc), sunlit background (Kg), shaded crown (Kt), and shaded background (Kz) are 0.853, 0.857, 0.914, and 0.838, respectively) compared with POV-Ray simulation, and performs better than GOMST, especially in scenes with high crown density. Moreover, SGOT outperformed the compared models in computational efficiency (4% faster than GOMST and 29.5% faster than GOST). Finally, the simulations of the scene components distribution in different topographic factors and crown density were further discussed. SGOT and GOST can both capture scene component variations caused by terrain better than GOMST, but comparatively, SGOT provides a more efficient tool to simulate the crown scene components because of its physical soundness and mathematical simplicity, and consequently, it will facilitate the modeling of canopy reflectance over mountainous regions.

Topography impacts the fraction of radiation that reaches a target surface from the sun, sky, and surrounding terrains, which leads to distortions in the radiance levels that are observed by sensors. Although several mountain radiative transfer (RT) models have been developed, the surrounding topographic effects are rarely considered, which can cause nonnegligible uncertainty in forward radiance modeling. In this study, we first extended the canopy reflectance model to be suitable for both continuous and discontinuous canopies over sloping terrain (BOST) to rugged terrain, and used it to investigate the impact of the surrounding topography on the reflected radiance and evaluate the contributions of diffuse irradiance from the sky and adjacent terrains. Discrete anisotropic radiative transfer (DART) simulations and remote sensing observations in a real mountain environment were used to evaluate the physical mechanism and model performance. The results suggested that the extended BOST model can capture terrain-induced variations in direct and diffuse radiation, which can be successfully used to simulate the reflected radiance in a real mountainous region. The model significantly improved compared with that before extension (i.e., RMSE decreases (R ² increases) of 1.252 (0.025) and 2.035 (0.054) in the red and NIR bands, respectively). In addition, we discovered that the surrounding topographic effects on the diffuse sky and terrain irradiance are significantly influenced by the wavelength, solar direction and visible area of the sky and terrains. The extended BOST model serves as an effective tool for improving simulations of the observed radiance and facilitates the development of RT models for rugged terrains.

Remote sensing change detection (RSCD) aims to identify changes between bi-temporal remote sensing images and plays a vital role in ecological monitoring, disaster response, urban planning, and climate change analysis. Existing RSCD methods often rely on simple feature differencing or stacking, which limits their ability to extract high-level discriminative change information. Furthermore, they typically neglect the interactions between difference features and low-level bi-temporal features, leading to suboptimal performance in complex scenes. To address these challenges, we propose MDENet, a Multi-scale Difference Information Guided Dual-temporal Enhancement Network, designed under an encoder–decoder architecture. MDENet incorporates four key modules: (1) the Difference Information Analysis Module (DIAM) for enhancing semantic difference features via complementary attention and convolutional pathways; (2) the Dual-temporal Difference Enhancement Module (DDEM) to strengthen the correlation between difference features and shallow bi-temporal features; (3) the Content-Guided Integration (CGI) module to spatially and contextually fuse change representations; and (4) the Difference Aggregation Module (DAM) to facilitate multi-scale difference interaction and semantic preservation. Extensive experiments on three datasets—LEVIR-CD, SYSU-CD, and our newly constructed CDTH-CD—demonstrate that MDENet consistently outperforms state-of-the-art RSCD methods, particularly under challenging conditions such as varying illumination and large-scale shadows. The code and dataset are available at: https://github.com/Jack-L23333/ MDENet.