Algorithm design and implementation for the detection of large herbivores from low-altitude (200 m–350 m) UAV remote sensing images faces two key problems: (1) the size of a single image from the UAV is too large, and the mainstream algorithm cannot adapt to it, and (2) the number of animals in the image is very small and densely distributed, which makes the model prone to missed detection. This paper proposes the following solutions: For the problem of animal size, we optimized the Faster-RCNN algorithm in terms of three aspects: selecting a HRNet feature extraction network that is more suitable for small target detection, using K-means clustering to obtain the anchor frame size that matches the experimental object, and using NMS to eliminate detection frames that have sizes inconsistent with the size range of the detection target after the algorithm generates the target detection frames. For image size, bisection segmentation was used when training the model, and when using the model to detect the whole image, we propose the use of a new overlapping segmentation detection method. The experimental results obtained for detecting yaks, Tibetan sheep (Tibetana folia), and the Tibetan wild ass in remote sensing images of low-altitude UAV from Maduo County, the source region of the Yellow River, show that the mean average precision (mAP) and average recall (AR) of the optimized Faster-RCNN algorithm are 97.2% and 98.2%, respectively, which are 9.5% and 12.1% higher than the values obtained by the original Faster-RCNN. In addition, the results obtained from applying the new overlap segmentation method to the whole UAV image detection process also show that the new overlap segmentation method can effectively solve the problems of the detection frames not fitting the target, missing detection, and creating false alarms due to bisection segmentation.

Planting suitability determines the distribution and yield of crops in a given region which can be greatly affected by climate change. In recent years, many studies have shown that carbon dioxide fertilization effects increase the productivity of temperate deciduous fruit trees under a changing climate, but the potential risks to fruit tree planting caused by a reduction in suitable planting areas are rarely reported. In this study, Maxent was first used to investigate the spatial distribution of five Pyrus species in China, and the consistency between the actual production area and the modeled climatically suitable area under the current climatic conditions were determined. In addition, based on Coupled Model Intercomparison Project Phase 6, three climate models were used to simulate the change in suitable area and the migration trend for different species under different emission scenarios (SSP1-2.6, SSP2-4.5, SSP3-7.0 and SSP5-8.5). The results showed that the suitable area for pear was highly consistent with the actual main production area under current climate conditions. The potential planting areas of P. ussuriensis showed a downward trend under all emission paths from 2020 to 2100; other species showed a trend of increasing first and then decreasing or slowing down and this growth effect was the most obvious in 2020–2040. Except for P. pashia, other species showed a migration trend toward a high latitude, and the trend was more prominent under the high emission path. Our results emphasize the response difference between species to climate change, and the method of consistency analysis between suitable planting area and actual production regions cannot only evaluate the potential planting risk but also provide a reasonable idea for the accuracy test of the modeled results. This work has certain guiding and reference significance for the protection of pear germplasm resources and the prediction of yield.

The carbon balance of terrestrial ecosystems is intertwined with climate and changes in land use. Over the past 30 years, the Loess Plateau (LP) has experienced temperature increases and an expansion of forest and grassland. The net ecosystem productivity (NEP) underlying these changes is worth investigating. Using three periods (i.e., 1990–2000, 2000–2010, and 2010–2019) of annual average NEP and climatic, topographic, and land use data, we analyzed changes in the spatial distribution of carbon sources and sinks of the LP. Using an optimal parameter-based geographical detector model to discuss the driving factors of carbon sources and sinks, we found that: (1) The area of carbon sinks has been increasing continuously, and that the distributions of both of these elements are zonal. The carbon sinks show a downward trend from south to north, which is mainly driven by climate and land use type. (2) Carbon sources are mainly concentrated in the middle temperate zone, and they are mainly linked to impervious land, unused land, and grassland. The carbon sinks are mainly concentrated in the south temperate zone and plateau climatic zone, and they are mainly linked to forest, grassland, and cultivated land. Additionally, the southern temperate zone has been the most green, due to its superior hydrothermal conditions that sustain carbon sinks. (3) It is not uncommon for some forests, grasslands, and cultivated land to transition between being carbon sources and carbon sinks, especially when affected by human intervention and inadequate management measures. (4) NEP was primarily influenced by CO₂ concentration, temperature, and precipitation, and the interaction of these factors greatly influenced the dynamics of carbon sources and sinks, while terrain exerted insignificant impacts on the NEP. This study highlights the importance of the carbon balance in terrestrial ecosystems and can be used to guide the creation of vegetation-based carbon sequestration policies.

The contribution of biodiversity to human well-being is key to exploring the relationships between biodiversity, ecosystem services (ES), and human well-being. In this work, a composite index, termed the human well-being index (HWI), was constructed for evaluating the contribution of biodiversity to human well-being in the Sanjiangyuan area. This index consists of material, ecological regulation, and spiritual and cultural contributions, represented by the material index (MI), the ecological regulation index (ERI), and the spiritual and cultural index (SCI), respectively. The system was further used to evaluate the spatiotemporal distribution of human well-being at the county level in 2000, 2010, and 2020. HWI increased steadily across Sanjiangyuan over the study period, especially in the western and northeastern counties; its center of gravity shifted in the northward direction. The MI increased (decreased) in the west and northeast (southeast); its center of gravity shifted in the northeast direction. All counties showing changes in the ERI were located in the eastern part of Sanjiangyuan. The center of gravity of ERI did not change significantly. The SCI increased steadily across the study area, but was high in the west and low in the east; the center of gravity shifted in the northwest direction. The study findings can contribute toward quantifying biodiversity contributions to human well-being and the formulation of biodiversity conservation policies.