Brent A. Murray

Proximally sensed laser scanning presents new opportunities for automated forest ecosystem data capture. However, a gap remains in deriving ecologically pertinent information, such as tree species, without additional ground data. Artificial intelligence approaches, particularly deep learning (DL), have shown promise towards automation. Progress has been limited by the lack of large, diverse, and, most importantly, openly available labelled single‐tree point cloud datasets. This has hindered both (1) the robustness of the DL models across varying data types (platforms and sensors) and (2) the ability to effectively track progress, thereby slowing the convergence towards best practice for species classification. To address the above limitations, we compiled the FOR‐species20K benchmark dataset, consisting of individual tree point clouds captured using proximally sensed laser scanning data from terrestrial (TLS), mobile (MLS) and drone laser scanning (ULS). Compiled collaboratively, the dataset includes data collected in forests mainly across Europe, covering Mediterranean, temperate and boreal biogeographic regions. It includes scattered tree data from other continents, totaling over 20,000 trees of 33 species and covering a wide range of tree sizes and forms. Alongside the release of FOR‐species20K, we benchmarked seven leading DL models for individual tree species classification, including both point cloud (PointNet++, MinkNet, MLP‐Mixer, DGCNNs) and multi‐view 2D‐based methods (SimpleView, DetailView, YOLOv5). 2D Image‐based models had, on average, higher overall accuracy (0.77) than 3D point cloud‐based models (0.72). Notably, the performance was consistently >0.8 across scanning platforms and sensors, offering versatility in deployment. The top‐scoring model, DetailView, demonstrated robustness to training data imbalances and effectively generalized across tree sizes. The FOR‐species20K dataset represents an important asset for developing and benchmarking DL models for individual tree species classification using proximally sensed laser scanning data. As such, it serves as a crucial foundation for future efforts to classify accurately and map tree species at various scales using laser scanning technology, as it provides the complete code base, dataset, and an initial baseline representative of the current state‐of‐the‐art of point cloud tree species classification methods.

Deep learning semantic segmentation algorithms have provided improved frameworks for the automated production of land use and land cover (LULC) maps, which significantly increases the frequency of map generation as well as consistency of production quality. In this research, a total of 28 different model variations were examined to improve the accuracy of LULC maps. The experiments were carried out using Landsat 5/7 or Landsat 8 satellite images with the North American land change monitoring system (NALCMS) labels. The performance of various convolutional neural networks and extension combinations were assessed, where Visual Geometry Group Network with an output stride of 4, and modified U-Net architecture, provided the best results. Additional expanded analysis of the generated LULC maps was also provided. Using a deep neural network, this work achieved 92.4% accuracy for 13 LULC classes within southern Manitoba representing a 15.8% improvement over published results for the NALCMS. Based on the large regions of interest, higher radiometric resolution of Landsat 8 data resulted in better overall accuracies (88.04%) compare to Landsat 5/7 (80.66%) for 16 LULC classes. This represents an 11.44% and 4.06% increase in overall accuracy compared to previously published NALCMS results, including larger land area and higher number of LULC classes incorporated into the models compared to other published LULC map automation methods.

Estimating surface water using satellite data can improve estimates and predictions for hydroelectric energy generation. The goal of this research is to identify the best threshold method and image band combination to detect seasonal surface water at the basin scale. This paper explores Otsu, Minimum Entropy, and average of Otsu-Minimum Entropy threshold algorithms using Sentinel-2 narrow near-infrared band 8A (N-NIR), shortwave infrared bands 11 and 12 (SWIR-11 or SWIR-12) images for the Assiniboine River Basin (ARB). The highest precision (0.87) of surface water detection was obtained in autumn using Minimum Entropy (N-NIR & SWIR-11) threshold values for the ARB, however, the Minimum Entropy also had poor detection of surface water (low recall). Use of the naïve Bayesian classifier did not improve detection of surface water compared to using threshold values alone, and the use of threshold values generated for one basin was not transferable to another basin. We concluded that threshold methods should be assessed seasonally, at a per-tile scale before combining surface water products at a watershed scale, and products should be evaluated using non-balanced statistical measures for watersheds with high land-to-water ratios. Accuracy assessment is hampered by validation data which represents static hydrologic conditions, not contemporary, seasonal conditions.