分类： 地球科学 >> 地理学 提交时间： 2023-02-07 合作期刊: 《干旱区科学》
摘要： Check dams are widely used on the Loess Plateau in China to control soil and water losses, develop agricultural land, and improve watershed ecology. Detailed information on the number and spatial distribution of check dams is critical for quantitatively evaluating hydrological and ecological effects and planning the construction of new dams. Thus, this study developed a check dam detection framework for broad areas from high-resolution remote sensing images using an ensemble approach of deep learning and geospatial analysis. First, we made a sample dataset of check dams using GaoFen-2 (GF-2) and Google Earth images. Next, we evaluated five popular deep-learning-based object detectors, including Faster R-CNN, You Only Look Once (version 3) (YOLOv3), Cascade R-CNN, YOLOX, and VarifocalNet (VFNet), to identify the best one for check dam detection. Finally, we analyzed the location characteristics of the check dams and used geographical constraints to optimize the detection results. Precision, recall, average precision at intersection over union (IoU) threshold of 0.50 (AP50), IoU threshold of 0.75 (AP75), and average value for 10 IoU thresholds ranging from 0.500.95 with a 0.05 step (AP5095), and inference time were used to evaluate model performance. All the five deep learning networks could identify check dams quickly and accurately, with AP5095, AP50, and AP75 values higher than 60.0%, 90.0%, and 70.0%, respectively, except for YOLOv3. The VFNet had the best performance, followed by YOLOX. The proposed framework was tested in the Yanhe River Basin and yielded promising results, with a recall rate of 87.0% for 521 check dams. Furthermore, the geographic analysis deleted about 50% of the false detection boxes, increasing the identification accuracy of check dams from 78.6% to 87.6%. Simultaneously, this framework recognized 568 recently constructed check dams and small check dams not recorded in the known check dam survey datasets. The extraction results will support efficient watershed management and guide future studies on soil erosion in the Loess Plateau.
分类： 地球科学 >> 地球科学史 提交时间： 2018-09-18 合作期刊: 《干旱区科学》
摘要： Soil acidification is a major global issue of sustainable development for ecosystems. The increasing soil acidity induced by excessive nitrogen (N) fertilization in farmlands has profoundly impacted the soil carbon dynamics. However, the way in which changes in soil pH regulating the soil carbon dynamics in a deep soil profile is still not well elucidated. In this study, through a 12-year field N fertilization experiment with three N fertilizer treatments (0, 120, and 240 kg N/(hm2•a)) in a dryland agroecosystem of China, we explored the soil pH changes over a soil profile up to a depth of 200 cm and determined the responses of soil organic carbon (SOC) and soil inorganic carbon (SIC) to the changed soil pH. Using a generalized additive model, we identified the soil depth intervals with the most powerful statistical relationships between changes in soil pH and soil carbon dynamics. Hierarchical responses of SOC and SIC dynamics to soil acidification were found. The results indicate that the changes in soil pH explained the SOC dynamics well by using a non-linear relationship at the soil depth of 0–80 cm (P=0.006), whereas the changes in soil pH were significantly linearly correlated with SIC dynamics at the 100–180 cm soil depth (P=0.015). After a long-term N fertilization in the experimental field, the soil pH value decreased in all three N fertilizer treatments. Furthermore, the declines in soil pH in the deep soil layer (100–200 cm) were significantly greater (P=0.035) than those in the upper soil layer (0–80 cm). These results indicate that soil acidification in the upper soil layer can transfer excess protons to the deep soil layer, and subsequently, the structural heterogeneous responses of SOC and SIC to soil acidification were identified because of different buffer capacities for the SOC and SIC. To better estimate the effects of soil acidification on soil carbon dynamics, we suggest that future investigations for soil acidification should be extended to a deeper soil depth, e.g., 200 cm.
点击量
点击量
下载量
评论
