Search

Many parts of the world face severe and prolonged drought conditions, stressing the sustainability of water resources and agriculture. Transitioning to water-efficient crops is one strategy that can help adapt to water scarcity. An emerging drought-tolerant crop of interest is chia (Salvia hispanica). Yet, no study has compared its large-scale water use dynamics to those of widely established crops across the globe. Here, we use satellite data over multiple years to assess the water use efficiency of chia, alfalfa, corn, and soybean globally. Results show that chia consumed 13-38% less water than alfalfa, corn, and soy and assimilated 14-20% more carbon per amount of water used. Substituting 10% of Southwest United States alfalfa cultivation with chia would save 184 million liters of water per growing season, equivalent to the annual water consumption of 1,300 households. Future research shall explore the economic, societal, and environmental ramifications of substituting alfalfa with chia in dry areas worldwide. These insights can guide decision-makers in promoting sustainable agriculture and water resource management.

Remote sensing has been used to analyse water quality since its inception. With recent breakthroughs in machine learning and satellite technology, researchers have begun to take a deeper look at the process as a viable alternative to traditional methods for water quality analysis. This study offers a systematic review of articles in which numerous researchers collectively analysed 26,160 water samples and developed 329 equations through various mathematical models that assess 13 distinct environmental parameters. The aim of this study is to better understand how researchers have addressed the challenges of creating suitable models that use remote sensing data and to identify trends or consensus on water quality parameters. This research delves deeper into the creation of these models to examine the various statistical approaches and their performance by investigating individual mathematical models in the literature. This research also explores the usage and application of various satellite bands for water quality monitoring. Key findings include a detailed examination of the spectral regions utilized by each model, reliance on band ratios, the impact of limited data, an assessment of the literature's strengths and limitations, and recommendations for future directions. [ABSTRACT FROM AUTHOR]

In the aftermath of natural disasters, timely and accurate building damage assessment is crucial for effective disaster response. This research introduces an innovative integrated ResNet-U-Net ensemble model that significantly enhances the accuracy and efficiency of damage detection from satellite imagery. The model combines the deep feature extraction capabilities of ResNet with the precise localization advantages of U-Net, supplemented by attention mechanisms to focus on critical damage indicators. Methodologically, the research employs a dataset of high-resolution satellite images from post-Hurricane Iota, processed through normalization and augmentation techniques. The model demonstrates exceptional accuracy, achieving 96.91% on an unbalanced dataset representative of real-world conditions, and 89.35% on a balanced dataset. This performance indicates a substantial improvement over existing approaches, offering a rapid and reliable tool for aiding disaster relief efforts. The proposed model's computational efficiency and generalizability underscore its potential to transform disaster management practices. Future work will explore the model's application to various disaster types and integration with real-time data processing systems. [ABSTRACT FROM AUTHOR]

Over the course of NASA's Dawn Discovery mission, the onboard framing camera mapped Ceres across a wide wavelength spectrum at varying polar science orbits and altitudes. With increasing resolution, the uniqueness of the 92 km wide, young Occator crater became evident. Its central cryovolcanic dome, Cerealia Tholus, and especially the associated bright carbonate and ammonium chloride deposits—named Cerealia Facula and the thinner, more dispersed Vinalia Faculae—are the surface expressions of a deep brine reservoir beneath Occator. Understandably, this made this crater the target for future sample return mission studies. The planning and preparation for this kind of mission require the characterization of potential landing sites based on the most accurate topography and orthorectified image data. In this work, we demonstrate the capabilities of the freely available and open-source USGS Integrated Software for Imagers and Spectrometers (ISIS 3) and Ames Stereo Pipeline (ASP 2.7) in creating high-quality image data products as well as stereophotogrammetric (SPG) and multi-view shape-from-shading (SfS) digital terrain models (DTMs) of the aforementioned spectroscopically challenging features. The main data products of our work are four new DTMs, including one SPG and one SfS DTM based on High-Altitude Mapping Orbit (HAMO) (CSH/CXJ) and one SPG and one SfS DTM based on Low-Altitude Mapping Orbit (LAMO) (CSL/CXL), along with selected Extended Mission Orbit 7 (XMO7) framing camera (FC) data. The SPG and SfS DTMs were calculated to a GSD of 1 and 0.5 px, corresponding to 136 m (HAMO SPG), 68 m (HAMO SfS), 34 m (LAMO SPG), and 17 m (LAMO SfS). Finally, we show that the SPG and SfS approaches we used yield consistent results even in the presence of high albedo differences and highlight how our new DTMs differ from those previously created and published by the German Aerospace Center (DLR) and the Jet Propulsion Laboratory (JPL). [ABSTRACT FROM AUTHOR]

Road extraction from satellite data is of great importance in various fields such as climate change, urban planning, forestry and sustainable development. In addition, fast and accurate road detection plays a critical role in disaster management and smart city applications, especially in emergency situations. In this context, U-net architecture provides an effective solution for tasks such as semantic segmentation and urban planning support. In this work, Edge U-net, a different adaptation of the U-net architecture, is used to map roads and streets and to detect changes over time. When the performance of the architecture is evaluated using Mean Intersection over Union (mIoU) and global accuracy metrics, superior results are obtained compared to other studies in the literature. In addition, the performance of the model was improved by applying transfer learning to the ImageNet dataset and various hyperparameter settings were performed. The results of the study show that path inferences are detected with 98.4% accuracy. These results show that Edge U-Net architecture and deep learning methods can be effectively used in road detection applications from satellite imagery. [ABSTRACT FROM AUTHOR]

Environmental assessment and management of dams are essential for understanding and mitigating their impact on downstream vegetation in arid and semi-arid regions. These regions pose challenges due to water scarcity and extreme climatic conditions. This study uses remote sensing technology to investigate the complex relationships between downstream vegetation changes before and after dam construction and various environmental factors, including land surface temperature, soil moisture, soil salinity, groundwater, runoff water, temperature, and precipitation. This study is notable for its comprehensive parameters and methods, providing unique insight into the direct and indirect impacts of dam construction on vegetation dynamics across diverse climatic regions. The examination encompasses four case studies, including Hali and Baish in Saudi Arabia and Lakes Argyle and Julius in Australia. This analysis utilizes satellite imagery and reanalyzed data, which allow consideration of a long time scale without any restrictions. A moderated mediation analysis explores the direct and indirect associations between downstream flows, environmental factors, and vegetation. Additionally, factor analysis using Principal Component Analysis (PCA) and Two-Step Clustering analysis is utilized to gain deeper insights into the underlying correlations and groupings within the dataset. The research findings highlight the significant roles of runoff, groundwater, soil salinity, and other environmental factors in influencing vegetation before and after dam construction in arid and semi-arid regions. Before dam construction in Hali and Baish, soil salinity, precipitation, and groundwater were key factors for vegetation, with runoff directly affecting arid areas. After construction, soil salinity and groundwater remained crucial, but runoff's direct impact ceased. In semi-arid regions like Lake Argyle and Julius, soil moisture, soil salinity, precipitation, and groundwater remained important, with runoff affecting vegetation cover. The study emphasizes the need to consider these factors, along with the use of remote sensing technology, in ecological and land management studies, noting that factors influencing vegetation dynamics changed significantly before and after dam construction.Highlights: Utilizing multiple satellite observations and reanalyzed datasets, we aim to overcome the limitations of insufficient ground measurements. These combined factors used in this study collectively underscore the contrasting outcomes in runoff's impact on NDVI. Before the dam construction, runoff influenced arid vegetation; post-construction, impact decreased. Before and after dam construction, factors influencing vegetation dynamics shifted significantly. [ABSTRACT FROM AUTHOR]

Road extraction is a sub-domain of remote sensing applications; it is a subject of extensive and ongoing research. The procedure of automatically extracting roads from satellite imagery encounters significant challenges due to the multi-scale and diverse structures of roads; improvement in this field is needed. Convolutional neural networks (CNNs), especially the DeepLab series known for its proficiency in semantic segmentation due to its efficiency in interpreting multi-scale objects' features, address some of these challenges caused by the varying nature of roads. The present work proposes the utilization of DeepLabV3+, the latest version of the DeepLab series, by introducing an innovative Dense Depthwise Dilated Separable Spatial Pyramid Pooling (DenseDDSSPP) module and integrating it in the place of the conventional Atrous Spatial Pyramid Pooling (ASPP) module. This modification enhances the extraction of complex road structures from satellite images. This study hypothesizes that the integration of DenseDDSSPP with a CNN backbone network and a Squeeze-and-Excitation block will generate an efficient dense feature map by focusing on relevant features, leading to more precise and accurate road extraction from remote sensing images. The Results Section presents a comparison of our model's performance against state-of-the-art models, demonstrating better results that highlight the effectiveness and success of the proposed approach. [ABSTRACT FROM AUTHOR]

On February 6, 2023, a major earthquake of 7.8 magnitude and its aftershocks caused widespread destruction in Turkey and Syria, causing more than 55,000 deaths, displacing 3 million people in Turkey and 2.9 million in Syria, and destroying or damaging at least 230,000 buildings. Our research presents detailed city-scale maps of landslides, liquefaction, and building damage from this earthquake, utilizing a novel variational causal Bayesian network. This network integrates InSAR-derived change detection with new empirical ground failure models and building footprints, enabling us to (1) rapidly estimate large-scale building damage, landslides, and liquefaction from remote sensing data, (2) jointly attribute building damage to landslides, liquefaction, and shaking, (3) improve regional landslide and liquefaction predictions impacting infrastructure, and (4) simultaneously identify damage degrees in thousands of buildings. For city-scale, building-by-building damage assessments, we use building footprints and satellite imagery with a spatial resolution of approximately 30 meters. This allows us to achieve a high resolution in damage assessment, both in timeliness and scale, enabling damage classification at the individual building level within days of the earthquake. Our findings detail the extent of building damage, including collapses, in Hatay, Osmaniye, Adıyaman, Gaziantep, and Kahramanmaras. We classified building damages into five categories: no damage, slight, moderate, partial collapse, and collapse. We evaluated damage estimates against preliminary ground-truth data reported by the civil authorities. Our results demonstrate the accuracy of our classification system, as evidenced by the area under the curve (AUC) scores on the receiver operating characteristic (ROC) curve, which ranged from 0.9588 to 0.9931 across different damage categories and regions. Specifically, our model achieved an AUC of 0.9931 for collapsed buildings in the Hatay/Osmaniye area, indicating a 99.31% probability that the model will rank a randomly chosen collapsed building higher than a randomly chosen non-collapsed building. These accurate, building-specific damage estimates, with greater than 95% classification accuracy across all categories, are crucial for disaster response and can aid agencies in effectively allocating resources and coordinating efforts during disaster recovery. [ABSTRACT FROM AUTHOR]

This study investigates the application of explainable AI (XAI) techniques to understand the deep learning models used for predicting urban conflict from satellite imagery. First, a ResNet18 convolutional neural network achieved 89% accuracy in distinguishing riot and non-riot urban areas. Using the Score-CAM technique, regions critical to the model's predictions were identified, and masking these areas caused a 20.9% drop in the classification accuracy, highlighting their importance. However, Score-CAM's ability to consistently localize key features was found to be limited, particularly in complex, multi-object urban environments. Analysis revealed minimal alignment between the model-identified features and traditional land use metrics, suggesting that deep learning captures unique patterns not represented in existing GIS datasets. These findings underscore the potential of deep learning to uncover previously unrecognized socio-spatial dynamics while revealing the need for improved interpretability methods. This work sets the stage for future research to enhance explainable AI techniques, bridging the gap between model performance and interpretability and advancing our understanding of urban conflict drivers. [ABSTRACT FROM AUTHOR]

This study presents an in-depth spatiotemporal analysis of mangrove ecosystems along Egypt's Red Sea coast, utilizing satellite imagery and GIS to examine changes from 2003 to 2022. We evaluate the effects of hydrological factors, specifically rainfall and runoff -presented by Stream Power Index-, on mangrove growth patterns. Results indicate a significant increase in mangrove areas, with a notable annual growth rate, despite a reduction in a specific region. This research highlights the integral role of catchment area runoff (R2 = 0.735, R = 0.857, P-value = 0.003 < 0.05, CV = 70.26%), rather than direct rainfall, in mangrove expansion, contributing to the understanding of mangrove resilience and informing sustainable coastal management strategies. The study bridges a significant research gap by mapping decadal mangrove changes, offering insights into the dynamics affecting these crucial ecosystems. [ABSTRACT FROM AUTHOR]

Land cover mapping is essential to understanding global land-use patterns and studying biodiversity composition and the functioning of eco-systems. The introduction of remote sensing technologies and artificial intelligence models made it possible to base land cover mapping on satellite imagery in order to monitor changes, assess ecosystem health, support conservation efforts, and reduce monitoring time. However, significant challenges remain in managing large, complex satellite imagery datasets, acquiring specialized datasets due to high costs and labor intensity, including a lack of comparative studies for the selection of optimal deep learning models. No less important is the scarcity of aerial datasets specifically tailored for agricultural areas. This study addresses these gaps by presenting a methodology for semantic segmentation of land covers in agricultural areas using satellite images and deep learning models with pre-trained backbones. We introduce an efficient methodology for preparing semantic segmentation datasets and contribute the "Land Cover Aerial Imagery" (LICAI) dataset for semantic segmentation. The study focuses on the Franciacorta area, Lombardy Region, leveraging the rich diversity of the dataset to effectively train and evaluate the models. We conducted a comparative study, using cutting-edge deep-learning-based segmentation models (U-Net, SegNet, DeepLabV3) with various pre-trained backbones (ResNet, Inception, DenseNet, EfficientNet) on our dataset acquired from Google Earth Pro. Through meticulous data acquisition, preprocessing, model selection, and evaluation, we demonstrate the effectiveness of these techniques in accurately identifying land cover classes. Integrating pre-trained feature extraction networks significantly improves performance across various metrics. Additionally, addressing challenges such as data availability, computational resources, and model interpretability is essential for advancing the field of remote sensing, in support of biodiversity conservation and the provision of ecosystem services and sustainable agriculture. [ABSTRACT FROM AUTHOR]

Leveraging satellite monitoring and machine learning (ML) techniques for water clarity assessment addresses the critical need for sustainable water management. This study aims to assess water clarity by predicting the Secchi disk depth (SDD) using satellite images and ML techniques. The primary methods involve data preparation and SSD inference. During data preparation, AquaSat samples, originally from the L1TP collection, were updated with the Landsat 8 satellite's latest postprocessing, L2SP, which includes atmospheric corrections, resulting in 33,261 multispectral observations and corresponding SSD measurements. For inferring the SSD, regressors such as SVR, NN, and XGB, along with an ensemble of them, were trained. The ensemble demonstrated performance with an average determination coefficient of R 2 of around 0.76 and a standard deviation of around 0.03. Field data validation achieved an R 2 of 0.80. Furthermore, we show that the regressors trained with L1TP imagery for predicting SSD result in a favorable performance with respect to their counterparts trained on the L2SP collection. This document contributes to the transition from semi-analytical to data-driven methods in water clarity research, using an ML ensemble to assess the clarity of water bodies through satellite imagery. [ABSTRACT FROM AUTHOR]

Cross-view geo-localization (CVGL) aims to determine the capture location of street-view images by matching them with corresponding 2D maps, such as satellite imagery. While recent bird's eye view (BEV)-based methods have advanced this task by addressing viewpoint and appearance differences, the existing approaches typically rely solely on either OpenStreetMap (OSM) data or satellite imagery, limiting localization robustness due to single-modality constraints. This paper presents a novel CVGL method that fuses OSM data with satellite imagery, leveraging their complementary strengths to enhance localization robustness. We integrate the semantic richness and structural information from OSM with the high-resolution visual details of satellite imagery, creating a unified 2D geospatial representation. Additionally, we employ a transformer-based BEV perception module that utilizes attention mechanisms to construct fine-grained BEV features from street-view images for matching with fused map features. Compared to state-of-the-art methods that utilize only OSM data, our approach achieves substantial improvements, with 12.05% and 12.06% recall enhancements on the KITTI benchmark for lateral and longitudinal localization within a 1-m error, respectively. [ABSTRACT FROM AUTHOR]

Recent statistics reveal a rapid decline in green spaces due to urbanization, impacting both urban and forested areas. Neglecting environmental concerns in urban planning has led to the need to restore greenery in communities. Hence, the objective of this research is to develop a predictive system using deep learning models to forecast changes in green spaces of a specific location over time, contributing to offering insights into urban planning and environmental conservation. Five time series model were explored to predict the change in green space of a location in future, with Long Short-Term Memory (LSTM) demonstrating superior performance. Furthermore, a two-module Siamese-LSTM framework was proposed to forecast future Green View Index (GVI) differences between consecutive years with an i-year gap. The proposed two-module Siamese-LSTM framework predicts the change in a location’s future green view index consisting of individual Siamese Models and an LSTM Model, achieving promising results with an MAE of 0.271, MSE of 0.150 and MAPE of 0.1989. [ABSTRACT FROM AUTHOR]