Small object detection presents significant challenges in computer vision, often affected by factors such as low resolution, dense object distribution, and complex backgrounds, which can lead to false positives or missed detections. In this paper, we introduce SRM-YOLO, a novel small object detection algorithm based on the YOLOv8 framework. The model incorporates the following key innovations: Reuse Fusion Structure (RFS), which enhances feature fusion; SPD-Conv, which enables effective downsampling while preserving critical information; and a specialized detection head designed for small objects. Additionally, the MPDIoU loss function is employed to improve detection accuracy. Experimental results on the VisDrone2019 dataset show that SRM-YOLO significantly enhances detection accuracy, achieving a 5.2% improvement in mAP50 over YOLOv8n. Additionally, its superior performance on the SSDD and NWPU VHR-10 datasets further validates its effectiveness in small object detection tasks.

Deep learning-based super-resolution (SR) techniques play a crucial role in enhancing the spatial resolution of images. However, remote sensing images present substantial challenges due to their diverse features, complex structures, and significant size variations in ground objects. Moreover, recovering lost details from low-resolution remote sensing images with complex and unknown degradations, such as downsampling, noise, and compression, remains a critical issue. To address these challenges, we propose ConvMambaSR, a novel super-resolution framework that integrates state-space models (SSMs) and Convolutional Neural Networks (CNNs). This framework is specifically designed to handle heterogeneous and complex ground features, as well as unknown degradations in remote sensing imagery. ConvMambaSR leverages SSMs to model global dependencies, activating more pixels in the super-resolution task. Concurrently, it employs CNNs to extract local detail features, enhancing the model’s ability to capture image textures and edges. Furthermore, we have developed a global–detail reconstruction module (GDRM) to integrate diverse levels of global and local information efficiently. We rigorously validated the proposed method on two distinct datasets, RSSCN7 and RSSRD-KQ, and benchmarked its performance against state-of-the-art SR models. Experiments show that our method achieves SOTA PSNR values of 26.06 and 24.29 on these datasets, respectively, and is visually superior, effectively addressing a variety of scenarios and significantly outperforming existing methods.

Identifying the relative contributions of climate change and human activities to alpine grassland dynamics is critical for understanding grassland degradation mechanisms. In this study, first, the actual NPP (NPP_a) was obtained by MOD17A3. Second, we used the Zhou Guangsheng model to simulate the potential met net primary productivity (NPP_p). Finally, the NPP generated by anthropogenic activities (NPP_h) was estimated by calculating the difference between NPP_p and NPP_a. Then, the relative contributions of climate change and human activities to NPP changes in grasslands were quantitatively assessed by analyzing trends in NPP_p and NPP_a. Thereby, the drivers of NPP change in the Yellow River source grassland were identified. The results showed that the temperature and precipitation in the study area showed a warm-humid climate trend from 2000 to 2020. The NPP_p and NPP_a increased at a rate of 1.07 g C/m² and 1.51 g C/m² per year, respectively, while the NPP_h decreased at a rate of 0.46 g C/m² per year. It can be seen that human activities had a positive effect on the change of NPP in the Yellow River source grassland from the change rate. The relative contribution analysis showed that 55.90% of grassland NPP increased due to climate change, 40.16% of grassland NPP increased due to human activities, and the grassland degradation was not significant. The research results can provide a theoretical basis and technical support for the next step of the Yellow River source grassland ecological protection project.

In view of the problem of sustainable restoration of vegetation in alpine mining areas, vegetation communities and physical and chemical properties of soil under different artificial restoration measures (i.e., grass monoculture, sowing quantity and topsoil replacement) were investigated for five consecutive years (2016–2020) in the sloped eastern area of a northern slag mound in the Jiangcang mining area of the Muli coalfield in Qinghai, China. The results showed that the vegetation characteristics of different sowing treatments with different kinds of grass species, such as Elymus nutans, Poa pratensis cv. Qinghai and Poa crymophila cv. Qinghai, were significantly different (p < 0.05). The content of soil available nitrogen and phosphorus was significantly different among different grass species (p < 0.05). Vegetation coverage and ramet density of sowing treatment five were significantly higher than those of a lower sowing quantity (p < 0.05). There was no significant difference in the vegetation characteristics among different grass species treatments five years after the restoration (p > 0.05). There was no significant difference in vegetation height of different soil covering treatments (p > 0.05). In the early stage of restoration, vegetation coverage and ramet density in TR2 (resurfacing soil 10 cm) and TR3 (resurfacing soil 15 cm) were significantly higher than those without treatment. Soil N, P and organic matter under the treatments of TR1 (resurfacing soil 5 cm), TR2 and TR3 were significantly higher than those in CK (p < 0.05) in the early stage of restoration, but there was no significant difference in soil N, P and organic matter after five years of restoration. Overall, the vegetation and soil characteristics showed a trend of increase first and then decrease during the 5-year restoration period under different artificial measures. There were significant differences in the vegetation and soil characteristics among different treatments in the second year of restoration (p < 0.05), but there was no significant difference between the first year and the fifth year of restoration, which indicated that vegetation and soil began to degrade after five years of restoration, and substrate nutrients in the mining waste soil could only support the short-term restoration of vegetation. Therefore, it is necessary to target matrix nutrients in future vegetation restoration in alpine mining areas.

Accurately assessing grazing impacts is essential for sustaining alpine grasslands. Conventional approaches often rely on total forage productivity, an indirect and uncertain proxy for forage availability. In this study, we propose a novel framework for estimating grazing pressure that integrates residual biomass with grazing intensity, thereby overcoming the limitations and uncertainties inherent in total forage-based assessments. Our results reveal pronounced spatiotemporal variation in grazing intensity: lowland areas experienced the highest intensity early in the growing season, whereas upland areas became more heavily grazed later in the season. However, grazing intensity alone proved insufficient to explain grazing pressure or predict pasture degradation risk. Overlay analyses demonstrated that only 38.8% of high intensity areas identified as under high grazing pressure, and more than 40% of high intensity area exhibiting substantial aboveground biomass. These findings highlight the limited explanatory power of grazing intensity when considered in isolation. By explicitly incorporating standing biomass rather than relying merely on total production, the proposed framework reduces estimation uncertainty, enhances ecological realism, and provides a scalable, more accurate and practical tool for monitoring grassland utilization and degradation.

This study investigated the effects of plateau pika (Ochotona curzoniae) disturbances and altitude on soil organic carbon (SOC) storage characteristics, including SOC content and SOC density (SOCD). In this study, plateau pika outbreak areas and non-outbreak areas at different altitudes were compared in terms of vegetation biomass, soil physicochemical properties, SOC content and SOCD to establish the relationship between vegetation and soil characteristics (including SOC content and SOCD). The results showed that SOC and SOCD decreased significantly (p < 0.01) in plateau pika outbreak areas, but SOCD increased first and then decreased with elevation in non-outbreak areas. Soil total nitrogen (TN) content decreased significantly (p < 0.01) with elevation in both plateau pika outbreak and non-outbreak areas. There were significant differences (p < 0.05) in total phosphorus (TP) at low elevations and nitrate nitrogen (NO₃⁻-N) at high elevations between outbreak and non-outbreak areas, but other nutrients did not differ hugely between outbreak and non-outbreak areas at the same elevation. Correlation analysis revealed that belowground biomass (BGB) in the plateau pika outbreak area was significantly and positively correlated with SOC (p < 0.01); structural equation modeling (SEM) analysis revealed that altitude had a direct effect on SOC (path coefficient = −0.882, p < 0.001) in the plateau pika outbreak area, but only a reduced influence on SOC and SOCD in the non-outbreak area; nitrate nitrogen in the plateau pika outbreak area and TN were the key influencing factors, which exerted a strong direct influence on SOC and SOCD (path coefficient = −0.666 and 0.639 (p < 0.001), respectively). Therefore, increasing vegetation biomass and nitrogen nutrient content through reseeding pasture and fertilization can facilitate the accumulation and recovery of SOC and SOCD in the ecological restoration of degraded alpine meadows, and it is especially important to quickly enrich soil nitrogen content in the outbreak area of plateau pika populations at high altitudes.