Search Results (2)

Water erosion is one of the major land degradation problems all over the globe, and its accurate quantification in different land use contexts is required in order to propose suitable conservation measures and curtail related hazards. In the Andaman and Nicobar (A&N) Islands, the land use changes due to faster urbanization and deforestation practices have led to accelerated erosion at many points around the inhabited Islands. Moreover, agricultural land uses in the A&N Islands are vulnerable to severe soil erosion, mainly due to cultivation practices along the steep slopes and mono-cropping culture. A study was conducted by establishing runoff plots in areas with different land uses to measure soil and nutrient losses and to estimate soil erosion using a semi-process-based soil erosion model, i.e., Revised Morgan Morgan and Finney (RMMF). The RMMF model was calibrated using primary data from runoff plots for the years 2019–21, validated for the year 2022, and applied in a Geographical Information System (GIS) to estimate soil erosion spatially over the Andaman ecosystem. The RMMF model simulated soil erosion during validation with a coefficient determination (R²) greater than 0.87 as compared to measured soil erosion from the runoff plots. The study revealed that annual N, P, and K losses of 41–81%, 42–95%, and 7–23%, respectively, due to runoff from various land uses. The land use land classification analysis of the Andaman Islands revealed that about 88% of the total geographical area is under the forest and mangrove land uses, which exhibited very slight soil erosion of <5 t/ha. This 88% of forest and mangrove areas requires suitable conservation measures such as afforestation and rehabilitation/restoration of mangroves. Moreover, 6% of cultivated areas need terracing, bunding, intercropping, etc., at the highest priority in order to conserve a sustainable Andaman ecosystem. On average, the annual soil loss from the Andaman Islands is 3.13 t/ha. About 6% of the study area exceeds the soil tolerance limit of 2.5–12.5 t/ha/year, which needs suitable soil and water conservation measures at the lowest priority due to economic implications. Full article

As the acquisition of very high resolution (VHR) images becomes easier, the complex characteristics of VHR images pose new challenges to traditional machine learning semantic segmentation methods. As an excellent convolutional neural network (CNN) structure, U-Net does not require manual intervention, and its high-precision features are widely used in image interpretation. However, as an end-to-end fully convolutional network, U-Net has not explored enough information from the full scale, and there is still room for improvement. In this study, we constructed an effective network module: residual module under a multisensory field (RMMF) to extract multiscale features of target and an attention mechanism to optimize feature information. RMMF uses parallel convolutional layers to learn features of different scales in the network and adds shortcut connections between stacked layers to construct residual blocks, combining low-level detailed information with high-level semantic information. RMMF is universal and extensible. The convolutional layer in the U-Net network is replaced with RMMF to improve the network structure. Additionally, the multiscale convolutional network was tested using RMMF on the Gaofen-2 data set and Potsdam data sets. Experiments show that compared to other technologies, this method has better performance in airborne and spaceborne images. Full article