A global population assessment of the Chinstrap penguin (Pygoscelis antarctica).

Using satellite imagery, drone imagery, and ground counts, we have assembled the first comprehensive global population assessment of Chinstrap penguins (Pygoscelis antarctica) at 3.42 (95th-percentile CI: [2.98, 4.00]) million breeding pairs across 375 extant colonies. Twenty-three previously known Chinstrap penguin colonies are found to be absent or extirpated. We identify five new colonies, and 21 additional colonies previously unreported and likely missed by previous surveys. Limited or imprecise historical data prohibit our assessment of population change at 35% of all Chinstrap penguin colonies. Of colonies for which a comparison can be made to historical counts in the 1980s, 45% have probably or certainly declined and 18% have probably or certainly increased. Several large colonies in the South Sandwich Islands, where conditions apparently remain favorable for Chinstrap penguins, cannot be assessed against a historical benchmark. Our population assessment provides a detailed baseline for quantifying future changes in Chinstrap penguin abundance, sheds new light on the environmental drivers of Chinstrap penguin population dynamics in Antarctica, and contributes to ongoing monitoring and conservation efforts at a time of climate change and concerns over declining krill abundance in the Southern Ocean.

Use of Very High Spatial Resolution Commercial Satellite Imagery and Deep Learning to Automatically Map Ice-Wedge Polygons across Tundra Vegetation Types.

We developed a high-throughput mapping workflow, which centers on deep learning (DL) convolutional neural network (CNN) algorithms on high-performance distributed computing resources, to automatically characterize ice-wedge polygons (IWPs) from sub-meter resolution commercial satellite imagery. We applied a region-based CNN object instance segmentation algorithm, namely the Mask R-CNN, to automatically detect and classify IWPs in North Slope of Alaska. The central goal of our study was to systematically expound the DLCNN model interoperability across varying tundra types (sedge, tussock sedge, and non-tussock sedge) and image scene complexities to refine the understanding of opportunities and challenges for regional-scale mapping applications. We corroborated quantitative error statistics along with detailed visual inspections to gauge the IWP detection accuracies. We found promising model performances (detection accuracies: 89% to 96% and classification accuracies: 94% to 97%) for all candidate image scenes with varying tundra types. The mapping workflow discerned the IWPs by exhibiting low absolute mean relative error (AMRE) values (0.17-0.23). Results further suggest the importance of increasing the variability of training samples when practicing transfer-learning strategy to map IWPs across heterogeneous tundra cover types. Overall, our findings demonstrate the robust performances of IWPs mapping workflow in multiple tundra landscapes.

Understanding the Effects of Optimal Combination of Spectral Bands on Deep Learning Model Predictions: A Case Study Based on Permafrost Tundra Landform Mapping Using High Resolution Multispectral Satellite Imagery.

Deep learning (DL) convolutional neural networks (CNNs) have been rapidly adapted in very high spatial resolution (VHSR) satellite image analysis. DLCNN-based computer visions (CV) applications primarily aim for everyday object detection from standard red, green, blue (RGB) imagery, while earth science remote sensing applications focus on geo object detection and classification from multispectral (MS) imagery. MS imagery includes RGB and narrow spectral channels from near- and/or middle-infrared regions of reflectance spectra. The central objective of this exploratory study is to understand to what degree MS band statistics govern DLCNN model predictions. We scaffold our analysis on a case study that uses Arctic tundra permafrost landform features called ice-wedge polygons (IWPs) as candidate geo objects. We choose Mask RCNN as the DLCNN architecture to detect IWPs from eight-band Worldview-02 VHSR satellite imagery. A systematic experiment was designed to understand the impact on choosing the optimal three-band combination in model prediction. We tasked five cohorts of three-band combinations coupled with statistical measures to gauge the spectral variability of input MS bands. The candidate scenes produced high model detection accuracies for the F1 score, ranging between 0.89 to 0.95, for two different band combinations (coastal blue, blue, green (1,2,3) and green, yellow, red (3,4,5)). The mapping workflow discerned the IWPs by exhibiting low random and systematic error in the order of 0.17-0.19 and 0.20-0.21, respectively, for band combinations (1,2,3). Results suggest that the prediction accuracy of the Mask-RCNN model is significantly influenced by the input MS bands. Overall, our findings accentuate the importance of considering the image statistics of input MS bands and careful selection of optimal bands for DLCNN predictions when DLCNN architectures are restricted to three spectral channels.

An analysis of enhanced tree trimming effectiveness on reducing power outages.

We evaluated the effectiveness of an enhanced tree trimming (ETT) program for its ability to reduce tree-related power outages, and thereby improve resilience, on an electric utility distribution system during storm events. Evaluations encompassed thirteen years of trimming (i.e. 2005-2017) data and were performed for both backbone and lateral utility lines. Backbones included all three phase lines between a substation and a faultable device whereas all other lines were considered laterals. The study site spanned the entire state of Connecticut, where the dominant vegetation is temperate deciduous forest. We controlled for variations in weather, tree cover, and wire type, by pairing ETT-treated zones with nearby untreated zones. ETT-treated conductors had storm outage rates that were 0.07-0.36 outages/km/year lower than untreated conductors or 35-180% lower than the service-area's average annual outage rate for untreated conductors. ETT-treatment was associated with lower outage rates for "minor" outage types (i.e., blown fuse, tripped recloser, etc.) but the treatment effect was not statistically significant for "major" outage types (damaged poles or wires). System-wide ETT application, for the approximately 27,000 km of conductors in the study area, was predicted to reduce annual storm-related outages by an average of 81-104 and 318-759 outages/year for backbone and lateral lines, respectively. Our study provided a robust empirical evaluation of ETT and also proposes a geospatial methodology that controls for variations in weather and environment.

Using Deep Learning to Identify Utility Poles with Crossarms and Estimate Their Locations from Google Street View Images.

Traditional methods of detecting and mapping utility poles are inefficient and costly because of the demand for visual interpretation with quality data sources or intense field inspection. The advent of deep learning for object detection provides an opportunity for detecting utility poles from side-view optical images. In this study, we proposed using a deep learning-based method for automatically mapping roadside utility poles with crossarms (UPCs) from Google Street View (GSV) images. The method combines the state-of-the-art DL object detection algorithm (i.e., the RetinaNet object detection algorithm) and a modified brute-force-based line-of-bearing (LOB, a LOB stands for the ray towards the location of the target [UPC at here] from the original location of the sensor [GSV mobile platform]) measurement method to estimate the locations of detected roadside UPCs from GSV. Experimental results indicate that: (1) both the average precision (AP) and the overall accuracy (OA) are around 0.78 when the intersection-over-union (IoU) threshold is greater than 0.3, based on the testing of 500 GSV images with a total number of 937 objects; and (2) around 2.6%, 47%, and 79% of estimated locations of utility poles are within 1 m, 5 m, and 10 m buffer zones, respectively, around the referenced locations of utility poles. In general, this study indicates that even in a complex background, most utility poles can be detected with the use of DL, and the LOB measurement method can estimate the locations of most UPCs.