More Like this

1. Building an Online Learning Module for Satellite Remote Sensing Applications in Hydrologic Science

   https://doi.org/10.3390/rs12183009  

   Maggioni, Viviana; Girotto, Manuela; Habib, Emad; Gallagher, Melissa A. (September 2020, Remote Sensing)

   null (Ed.)

   This article presents an online teaching tool that introduces students to basic concepts of remote sensing and its applications in hydrology. The learning module is intended for junior/senior undergraduate students or junior graduate students with no (or little) prior experience in remote sensing, but with some basic background of environmental science, hydrology, statistics, and programming. This e-learning environment offers background content on the fundamentals of remote sensing, but also integrates a set of existing online tools for visualization and analysis of satellite observations. Specifically, students are introduced to a variety of satellite products and techniques that can be used to monitor and analyze changes in the hydrological cycle. At completion of the module, students are able to visualize remote sensing data (both in terms of time series and spatial maps), detect temporal trends, interpret satellite images, and assess errors and uncertainties in a remote sensing product. Students are given the opportunity to check their understanding as they progress through the module and also tackle complex real-life problems using remote sensing observations that professionals and scientists commonly use in practice. The learning tool is implemented in HydroLearn, an open-source, online platform for instructors to find and share learning modules and collaborate on developing teaching resources in hydrology and water resources. 

   more » « less
2. Within‐field soil moisture variability and time‐invariant spatial structures of agricultural fields in the US Midwest

   https://doi.org/10.1002/vzj2.20337  

   Yang, Yi; Peng, Bin; Guan, Kaiyu; Pan, Ming; Franz, Trenton_E; Cosh, Michael_H; Bernacchi, Carl_J (April 2024, Vadose Zone Journal)

   Abstract Understanding soil moisture variability and estimating high‐resolution soil moisture at subfield to field scales is critical for agricultural research and applications. However, systematic investigation of subfield scale soil moisture variability over cropland is still lacking from both measurement and satellite remote sensing. In this study, we aim to investigate (1) the characteristics of within‐field soil moisture distribution over typical cropland in the US Midwest and (2) the capabilities of satellite remote sensing in capturing the spatiotemporal variabilities of soil moisture at subfield scale. Specifically, we conducted soil moisture field experiments in three typical commercial agricultural fields (∼85 acres per field) in central Illinois, representing typical commercial farmlands in the US Midwest, and compared the soil moisture measurements with satellite remote sensing data from optical and active microwave sensors. In each field, dense soil moisture samples (spaced at 50–60 m) were obtained for two dry down events in May and July 2021, and multiple long‐term soil moisture stations were installed. We found prominent time‐invariant spatial structures of soil moisture at within‐field scales both during the dry down period and over longer time scales, and the stability is minimally affected by plant water use during the growing season. Comparing the field campaign measurements with satellite remote sensing data, we found that surface reflectance of shortwave infrared bands, such as SWIR1 (1610 nm) from Sentinel‐2, can capture relative surface soil moisture patterns at within‐field scales, but their relationships with soil moisture are field specific. These findings and the improved understanding of within‐field soil moisture dynamics could potentially help future research on high‐resolution soil moisture estimation with multi‐source remote sensing data. 

   more » « less
3. Assessing the availability of green infrastructure to residents of an Arctic city (on the example of Nadym) [Оценка обеспеченности зеленой инфраструктурой жителей арк­тического города (на примере Надыма)]

   https://doi.org/10.25283/2223-4594-2022-4-475-490  

   Sizov, O.S.; Fedorov, R.Y.; Pechkina, Y.A.; Michugin, M.S.; Kuklina, V.V.; Soromotin, A.V.; Fedash, A.V. (December 2022, Arctic: Ecology and Economy)

   Based on remote sensing data, the authors consider the features of the formation and use of green spaces in the city of Nadym (Yamalo-Nenets Autonomous Area). They give a detailed assessment of the availability of green infrastructure for the inhabitants of Nadym based on a comparison of the spatial distribution of vegetation and the urban population. During the construction of the city, there was a dramatic reduction in the area of vegetation cover, which reached its maximum during active construction in the 1980s. After the completion of the main construction stage and until now, there has been a steady increase in the share of vegetation, which is explained by active landscaping activities against the backdrop of climate softening. The authors have find out that while maintaining the high availability of open green spaces within the city, the main lack of vegetation is observed within the residential development of microdistricts. The methodology for the integrated use of medium and ultra-high resolution space images, UAV surveys, detailed mapping of residential buildings and field geobotanical descriptions tested during the study can be used in a detailed analysis of the state of the green infrastructure of other cities in the north of Western Siberia. In general, the assessment of the green infrastructure availability in the Arctic cities is of great importance for urban planning, allowing to fully take into account the regional environmental needs of local residents, in the context of the heterogeneity of their distribution. 

   more » « less
4. Wildland Fire Detection and Monitoring Using a Drone-Collected RGB/IR Image Dataset

   Chen, Xiwen and (January 2022, IEEE Access)

   Current forest monitoring technologies including satellite remote sensing, manned/piloted aircraft, and observation towers leave uncertainties about a wildfire’s extent, behavior, and conditions in the fire’s near environment, particularly during its early growth. Rapid mapping and real-time fire monitoring can inform in-time intervention or management solutions to maximize beneficial fire outcomes. Drone systems’ unique features of 3D mobility, low flight altitude, and fast and easy deployment make them a valuable tool for early detection and assessment of wildland fires, especially in remote forests that are not easily accessible by ground vehicles. In addition, the lack of abundant, well-annotated aerial datasets – in part due to unmanned aerial vehicles’ (UAVs’) flight restrictions during prescribed burns and wildfires – has limited research advances in reliable data-driven fire detection and modeling techniques. While existing wildland fire datasets often include either color or thermal fire images, here we present (1) a multi-modal UAV-collected dataset of dual-feed side-by-side videos including both RGB and thermal images of a prescribed fire in an open canopy pine forest in Northern Arizona and (2) a deep learning-based methodology for detecting fire and smoke pixels at accuracy much higher than the usual single-channel video feeds. The collected images are labeled to “fire” or “no-fire” frames by two human experts using side-by-side RGB and thermal images to determine the label. To provide context to the main dataset’s aerial imagery, the included supplementary dataset provides a georeferenced pre-burn point cloud, an RGB orthomosaic, weather information, a burn plan, and other burn information. By using and expanding on this guide dataset, research can develop new data-driven fire detection, fire segmentation, and fire modeling techniques. 

   more » « less
5. Hyperparameter Optimization for Large-Scale Remote Sensing Image Analysis Tasks: A Case Study Based on Permafrost Landform Detection Using Deep Learning

   https://doi.org/10.1109/ACCESS.2024.3379142  

   Perera, Amal S; Witharana, Chandi; Manos, Elias; Liljedahl, Anna K (January 2024, IEEE Access)

   Climate change pressure on the Arctic permafrost is rising alarmingly, creating a decisive need to produce Pan-Arctic scale permafrost landform and thaw disturbance information from remote sensing (RS) data. Very high spatial resolution (VHSR) satellite images can be utilized to detect ice-wedge polygons (IWPs) – the most important and widespread landform in the Arctic tundra region - across the Arctic without compromising spatial details. Automated analysis of peta-byte scale VHSR imagery covering millions of square kilometers is a computationally challenging task. Traditional semantic segmentation requires the use of task specific feature extraction with conventional classification techniques. Semantic complexity of VHSR images coupled with landscape heterogeneity makes it difficult to use conventional classification approaches to produce Pan-Arctic scale geospatial products. This leads to adapting deep convolutional neural network (DLCNN) approaches that have excelled in computer vision (CV) applications. Transitioning domains from everyday image understanding to remote sensing image analysis is challenging. This study aims to systematically investigate two main obstacles confronted when adapting DLCNNs in large-scale RS image analysis tasks; 1) the limited availability labeled data sets and 2) the prohibitive nature of hyperparameter tunning when designing DLCNNs that can capture the rich characteristics embedded in remotely-sensed images. With a case study on the production of the first pan-Arctic ice-wedge polygon map using thousands of VHSR images, we demonstrate the use of transfer learning and the impact of hyperparameter tuning with a 16% improvement of the Mean Average Precision (mAP50). 

   more » « less