Swin Transformer and Deep Convolutional Neural Networks for Coastal Wetland Classification Using Sentinel-1, Sentinel-2, and LiDAR Data

The use of machine learning algorithms to classify complex landscapes has been revolutionized by the introduction of deep learning techniques, particularly in remote sensing. Convolutional neural networks (CNNs) have shown great success in the classification of complex high-dimensional remote sensing imagery, specifically in wetland classification. On the other hand, the state-of-the-art natural language processing (NLP) algorithms are transformers. Although the transformers have been studied for a few remote sensing applications, the integration of deep CNNs and transformers has not been studied, particularly in wetland mapping. As such, in this study, we explore the potential and possible limitations to be overcome regarding the use of a multi-model deep learning network with the integration of a modified version of the well-known deep CNN network of VGG-16, a 3D CNN network, and Swin transformer for complex coastal wetland classification. Moreover, we discuss the potential and limitation of the proposed multi-model technique over several solo models, including a random forest (RF), support vector machine (SVM), VGG-16, 3D CNN, and Swin transformer in the pilot site of Saint John city located in New Brunswick, Canada. In terms of F-1 score, the multi-model network obtained values of 0.87, 0.88, 0.89, 0.91, 0.93, 0.93, and 0.93 for the recognition of shrub wetland, fen, bog, aquatic bed, coastal marsh, forested wetland, and freshwater marsh, respectively. The results suggest that the multi-model network is superior to other solo classifiers from 3.36% to 33.35% in terms of average accuracy. Results achieved in this study suggest the high potential for integrating and using CNN networks with the cutting-edge transformers for the classification of complex landscapes in remote sensing.

Why highways remake hierarchies

This chapter lays out the volume’s documentation of many of the uneven – and unexpected – experiences of mobility transformation as it unfolds as a developmental imperative across vast and complex landscapes of South Asia. Whether journeys become shorter, faster, more treacherous, cheaper, or more costly, questions about ownership, management, access to ‘public goods’, responsibility, and other critical concerns consistently take new shape when expressed through the coming of a new road or transportation network. We posit that roads are fragile political achievements. In response to the sweeping state promises about new mobilities and modernization that highways are purported to deliver, the stories comprising this volume, and outlined in this chapter, speak from other perspectives, such as how political opportunity is routinely met with a measure of public scepticism and at times efficacious protest.

Performance Comparison of Filtering Algorithms for High-Density Airborne LiDAR Point Clouds over Complex LandScapes

Airborne light detection and ranging (LiDAR) technology has become the mainstream data source in geosciences and environmental sciences. Point cloud filtering is a prerequisite for almost all LiDAR-based applications. However, it is challenging to select a suitable filtering algorithm for handling high-density point clouds over complex landscapes. Therefore, to determine an appropriate filter on a specific environment, this paper comparatively assessed the performance of five representative filtering algorithms on six study sites with different terrain characteristics, where three plots are located in urban areas and three in forest areas. The representative filtering methods include simple morphological filter (SMRF), multiresolution hierarchical filter (MHF), slope-based filter (SBF), progressive TIN densification (PTD) and segmentation-based filter (SegBF). Results demonstrate that SMRF performs the best in urban areas, and compared to MHF, SBF, PTD and SegBF, the total error of SMRF is reduced by 1.38%, 48.21%, 48.25% and 31.03%, respectively. MHF outperforms the others in forest areas, and compared to SMRF, SBF, PTD and SegBF, the total error of MHF is reduced by 1.98%, 35.87%, 45.11% and 9.42%, respectively. Moreover, both SMRF and MHF keep a good balance between type I and II errors, which makes the produced DEMs much similar to the references. Overall, SMRF and MHF are recommended for urban and forest areas, respectively, and MHF averagely performs slightly better than SMRF on all areas with respect to kappa coefficient.

Tracking the invasive hornet Vespa velutina in complex environments by means of a harmonic radar

An innovative scanning harmonic radar has been recently developed for tracking insects in complex landscapes. This movable technology has been tested on an invasive hornet species (Vespa velutina) for detecting the position of their nests in the environment, in the framework of an early detection strategy. The new model of harmonic radar proved to be effective in tracking hornets either in open landscapes, hilly environments and areas characterised by the presence of more obstacles, such as woodlands and urban areas. Hornets were effectively tracked in complex landscapes for a mean tracking length of 96 ± 62 m with maximum values of ~ 300 m. The effectiveness of locating nests was 75% in new invasive outbreaks and 60% in highly density colonised areas. Furthermore, this technology could provide information on several aspects of insect’s ecology and biology. In this case, new insights were obtained about the mean foraging range of V. velutina (395 ± 208 m with a maximum value of 786 m) and flying features (ground speed), which was 6.66 ± 2.31 m s−1 for foraging individuals (hornets that are not carrying prey’s pellet) and 4.06 ± 1.34 m s−1 for homing individuals.

Increasing connectivity enhances habitat specialists but simplifies plant–insect food webs

Strong declines of grassland species diversity in small and isolated grassland patches have been observed at local and landscape scales. Here, we study how plant–herbivore interaction webs and habitat specialisation of leafhopper communities change with the size of calcareous grassland fragments and landscape connectivity. We surveyed leafhoppers and plants on 14 small (0.1–0.6 ha) and 14 large (1.2–8.8 ha) semi-natural calcareous grassland fragments in Central Germany, differing in isolation from other calcareous grasslands and in the percentage of arable land in the surrounding landscape (from simple to complex landscapes). We quantified weighted trophic links between plants and their phytophagous leafhoppers for each grassland fragment. We found that large and well-connected grassland fragments harboured a high portion of specialist leafhopper species, which in turn yielded low interaction diversity and simple plant-leafhopper food webs. In contrast, small and well-connected fragments exhibited high levels of generalism, leading to higher interaction diversity. In conclusion, food web complexity appeared to be a poor indicator for the management of insect diversity, as it is driven by specialist species, which require high connectivity of large fragments in complex landscapes. We conclude that habitat specialists should be prioritized since generalist species associated with small fragments are also widespread in the surrounding landscape matrix.