Search

To date, no method utilizing satellite imagery exists for detailing the locations and functions of critical infrastructure across the United States, making response to natural disasters and other events challenging due to complex infrastructural interdependencies. This paper presents a repeatable, transferable, and explainable method for critical infrastructure analysis and implementation of a robust model for critical infrastructure detection in satellite imagery. This model consists of a DenseNet-161 convolutional neural network, pretrained with the ImageNet database. The model was provided additional training with a custom dataset, containing nine infrastructure classes. The resultant analysis achieved an overall accuracy of 90%, with the highest accuracy for airports (97%), hydroelectric dams (96%), solar farms (94%), substations (91%), potable water tanks (93%), and hospitals (93%). Critical infrastructure types with relatively low accuracy are likely influenced by data commonality between similar infrastructure components for petroleum terminals (86%), water treatment plants (78%), and natural gas generation (78%). Local interpretable model-agnostic explanations (LIME) was integrated into the overall modeling pipeline to establish trust for users in critical infrastructure applications. The results demonstrate the effectiveness of a convolutional neural network approach for critical infrastructure identification, with higher than 90% accuracy in identifying six of the critical infrastructure facility types. [ABSTRACT FROM AUTHOR]

The rising severity and frequency of wildfires in recent years in the United States have raised numerous concerns regarding the improvement in wildfire emergency response management and decision-making systems, which require operational high temporal and spatial resolution monitoring capabilities. Satellites are one of the tools that can be used for wildfire monitoring. However, none of the currently available satellite systems provide both high temporal and spatial resolution. For example, GOES-17 geostationary satellite fire products have high temporal (1–5 min) but low spatial resolution (≥2 km), and VIIRS polar orbiter satellite fire products have low temporal (~12 h) but high spatial resolution (375 m). This work aims to leverage currently available satellite data sources, such as GOES and VIIRS, along with deep learning (DL) advances to achieve an operational high-resolution, both spatially and temporarily, wildfire monitoring tool. Specifically, this study considers the problem of increasing the spatial resolution of high temporal but low spatial resolution GOES-17 data products using low temporal but high spatial resolution VIIRS data products. The main idea is using an Autoencoder DL model to learn how to map GOES-17 geostationary low spatial resolution satellite images to VIIRS polar orbiter high spatial resolution satellite images. In this context, several loss functions and DL architectures are implemented and tested to predict both the fire area and the corresponding brightness temperature. These models are trained and tested on wildfire sites from 2019 to 2021 in the western U.S. The results indicate that DL models can improve the spatial resolution of GOES-17 images, leading to images that mimic the spatial resolution of VIIRS images. Combined with GOES-17 higher temporal resolution, the DL model can provide high-resolution near-real-time wildfire monitoring capability as well as semi-continuous wildfire progression maps. [ABSTRACT FROM AUTHOR]

Regression algorithms are regularly used for improving the accuracy of satellite precipitation products. In this context, satellite precipitation and topography data are the predictor variables, and gauged-measured precipitation data are the dependent variables. Alongside this, it is increasingly recognised in many fields that combinations of algorithms through ensemble learning can lead to substantial predictive performance improvements. Still, a sufficient number of ensemble learners for improving the accuracy of satellite precipitation products and their large-scale comparison are currently missing from the literature. In this study, we work towards filling in this specific gap by proposing 11 new ensemble learners in the field and by extensively comparing them. We apply the ensemble learners to monthly data from the PERSIANN (Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks) and IMERG (Integrated Multi-satellitE Retrievals for GPM) gridded datasets that span over a 15-year period and over the entire contiguous United States (CONUS). We also use gauge-measured precipitation data from the Global Historical Climatology Network monthly database, version 2 (GHCNm). The ensemble learners combine the predictions of six machine learning regression algorithms (base learners), namely the multivariate adaptive regression splines (MARS), multivariate adaptive polynomial splines (poly-MARS), random forests (RF), gradient boosting machines (GBM), extreme gradient boosting (XGBoost) and Bayesian regularized neural networks (BRNN), and each of them is based on a different combiner. The combiners include the equal-weight combiner, the median combiner, two best learners and seven variants of a sophisticated stacking method. The latter stacks a regression algorithm on top of the base learners to combine their independent predictions. Its seven variants are defined by seven different regression algorithms, specifically the linear regression (LR) algorithm and the six algorithms also used as base learners. The results suggest that sophisticated stacking performs significantly better than the base learners, especially when applied using the LR algorithm. It also beats the simpler combination methods. [ABSTRACT FROM AUTHOR]

Severe forest disturbance events are becoming more common due to climate change and many forest managers rely heavily upon airborne surveys to map damage. However, when the damage is extensive, airborne assets are in high demand and it can take managers several weeks to account for the damage, delaying important management actions. While some satellite-based systems exist to help with this process, their spatial resolution or latency can be too large for the needs of managers, as evidenced by the continued use of airborne imaging. Here, we present a new, operational-focused system capable of leveraging high spatial and temporal resolution Sentinel-2 and Planet Dove imagery to support the mapping process. This system, which we have named Astrape ("ah-STRAH-pee"), uses recently developed techniques in image segmentation and machine learning to produce maps of damage in different forest types and regions without requiring ground data, greatly reducing the need for potentially dangerous airborne surveys and ground sampling needed to accurately quantify severe damage. Although some limited field work is required to verify results, similar to current operational systems, Astrape-produced maps achieved 78–86% accuracy with respect to damage severity when evaluated against reference data. We present the Astrape framework and demonstrate its flexibility and potential with four case studies depicting four different disturbance types—fire, hurricane, derecho and tornado—in three disparate regions of the United States. Astrape is capable of leveraging various sources of satellite imagery and offers an efficient, flexible and economical option for mapping severe damage in forests. [ABSTRACT FROM AUTHOR]

A task for environmental health research is to produce complete pollution exposure maps despite limited monitoring data. Satellite-derived aerosol optical depth (AOD) is frequently used as a predictor in various models to improve PM2.5 estimation, despite significant gaps in coverage. We analyze PM2.5 and AOD from July 2011 in the contiguous United States. We examine two methods to aid in gap-filling AOD: (1) lattice kriging, a spatial statistical method adapted to handle large amounts data, and (2) random forest, a tree-based machine learning method. First, we evaluate each model's performance in the spatial prediction of AOD, and we additionally consider ensemble methods for combining the predictors. In order to accurately assess the predictive performance of these methods, we construct spatially clustered holdouts to mimic the observed patterns of missing data. Finally, we assess whether gap-filling AOD through one of the proposed ensemble methods can improve prediction of PM2.5 in a random forest model. Our results suggest that ensemble methods of combining lattice kriging and random forest can improve AOD gap-filling. Based on summary metrics of performance, PM2.5 predictions based on random forest models were largely similar regardless of the inclusion of gap-filled AOD, but there was some variability in daily model predictions. [ABSTRACT FROM AUTHOR]

Vegetation water content (VWC) is recognized as an important parameter in vegetation growth studies, natural disasters such as forest fires, and drought prediction. Recently, the Global Navigation Satellite System Interferometric Reflectometry (GNSS-IR) has emerged as an important technique for monitoring vegetation information. The normalized microwave reflection index (NMRI) was developed to reflect the change of VWC based on this fact. However, NMRI uses local site-based data, and the sparse distribution hinders the application of NMRI. In this study, we obtained a 500 m spatially continuous NMRI product by integrating GNSS-IR site data with other VWC-related products using the point–surface fusion technique. The auxiliary data in the fusion process include the normalized difference vegetation index (NDVI), gross primary productivity (GPP), and precipitation. Meanwhile, the fusion performance of three machine learning methods, i.e., the back-propagation neural network (BPNN), generalized regression neural network (GRNN), and random forest (RF) are compared and analyzed. The machine learning methods achieve satisfactory results, with cross-validation R values of 0.71–0.83 and RMSEs of 0.025–0.037. The results show a clear improvement over the traditional multiple linear regression method, which achieves R (RMSE) values of only about 0.4 (0.045). It indicates that the machine learning methods can better learn the complex nonlinear relationship between NMRI and the input VWC-related index. Among the machine learning methods, the RF model obtained the best results. Long time-series NMRI images with a 500 m spatial resolution in the western part of the continental U.S. were then obtained. The results show that the spatial distribution of the NMRI product is consistent with a drought situation from 2012 to 2014 in the U.S., which verifies the feasibility of analyzing and predicting drought times and distribution ranges by using the 500 m fusion product. [ABSTRACT FROM AUTHOR]

Despite the need for quality land cover information, large-area, high spatial resolution land cover mapping has proven to be a difficult task for a variety of reasons including large data volumes, complexity of developing training and validation datasets, data availability, and heterogeneity in data and landscape conditions. We investigate the use of geographic object-based image analysis (GEOBIA), random forest (RF) machine learning, and National Agriculture Imagery Program (NAIP) orthophotography for mapping general land cover across the entire state of West Virginia, USA, an area of roughly 62,000 km2. We obtained an overall accuracy of 96.7% and a Kappa statistic of 0.886 using a combination of NAIP orthophotography and ancillary data. Despite the high overall classification accuracy, some classes were difficult to differentiate, as highlight by the low user's and producer's accuracies for the barren, impervious, and mixed developed classes. In contrast, forest, low vegetation, and water were generally mapped with accuracy. The inclusion of ancillary data and first- and second-order textural measures generally improved classification accuracy whereas band indices and object geometric measures were less valuable. Including super-object attributes improved the classification slightly; however, this increased the computational time and complexity. From the findings of this research and previous studies, recommendations are provided for mapping large spatial extents. [ABSTRACT FROM AUTHOR]

Satellite-derived estimates of aerosol optical depth (AOD) are key predictors in particulate air pollution models. The multi-step retrieval algorithms that estimate AOD also produce quality control variables but these have not been systematically used to address the measurement error in AOD. We compare three machine-learning methods: random forests, gradient boosting, and extreme gradient boosting (XGBoost) to characterize and correct measurement error in the Multi-Angle Implementation of Atmospheric Correction (MAIAC) 1 × 1 km AOD product for Aqua and Terra satellites across the Northeastern/Mid-Atlantic USA versus collocated measures from 79 ground-based AERONET stations over 14 years. Models included 52 quality control, land use, meteorology, and spatially-derived features. Variable importance measures suggest relative azimuth, AOD uncertainty, and the AOD difference in 30–210 km moving windows are among the most important features for predicting measurement error. XGBoost outperformed the other machine-learning approaches, decreasing the root mean squared error in withheld testing data by 43% and 44% for Aqua and Terra. After correction using XGBoost, the correlation of collocated AOD and daily PM2.5 monitors across the region increased by 10 and 9 percentage points for Aqua and Terra. We demonstrate how machine learning with quality control and spatial features substantially improves satellite-derived AOD products for air pollution modeling. [ABSTRACT FROM AUTHOR]