GeeViz is a free, open-source Python package designed for exploring and analyzing earth observation data on the Google Earth Engine (GEE) platform. GeeViz simplifies complex data processing, analysis, and visualization tasks with user-friendly functions, making it easier to work with GEE data. Before 2017, most of our GEE scripts were written in JavaScript. However, we realized that the Python API offered greater versatility but there was no equivalent to the Code Editor to debug our outputs. To address this, we developed a viewer framework that evolved into GeeViz, providing a powerful, accessible toolkit for anyone working with GEE.

Storing, hosting, and sharing imagery and raster data can be costly. In a collaborative effort with multiple agencies we developed a cloud solution for storing and delivering petabytes of data seamlessly to users. This solution simultaneously reduced costs and provided a scalable solution that can dynamically accommodate up to 50,000 users while reducing hosting and overhead costs. These streaming services simplify access to imagery, improve experience, and allow users to more easily utilize the best data available to make more informed decisions.

As a result of our successful use of Google Earth Engine (GEE) in the Landscape Change Modeling System, Google hired Redcastle Resources to develop the first-ever geospatial training for Google Cloud Skills Boost. We delivered this training over two days, in person, at GEE's Geo For Good 2023 conference. The content is freely available to the public. In the workshop, we simplified a complex geospatial processing workflow by translating it into a series of Python notebooks. Additionally, we created and delivered a series of lectures to an international audience. The focus of this workshop was on using the geeViz package to develop a continental-scale model of land cover, land change, and land use.

TreeMap, produced by the Fire Lab of the USFS Rocky Mountain Research Station, is a CONUS-wide map of forest biomass and carbon attributes. We were tasked to develop a visually appealing data explorer and produce new, annual layers of the dataset. We developed an interactive data viewer and download tool with standardized symbology and custom area summaries. We uploaded the dataset to the Google Earth Engine Data Catalog, where it can be freely accessed and used in cloud computing workflows by anyone in the world. We re-vamped the production process from an unordered collection of scripts into a modular, version-controlled set of functions that follow software development best practices. The resulting dataset is used in carbon-monitoring and wildfire crisis strategy analyses.

Following a catastrophic wildfire in New Mexico, monsoonal debris flows clogged stream channels and destroyed infrastructure. While scoping the project, we worked closely to understand the data needs of our customer and select the sensor type and UAS platform most appropriate for the project and site. By comparing high-resolution UAS lidar-measured elevation products with pre-event lidar sources, RedCastle was able to estimate the depth (and volume) of eroded and deposited sediment with <10 cm vertical accuracy. Our analysts’ expertise in every step of the lidar processing pipeline and their familiarity with other remote sensing data sources allowed us to deliver on this project, providing change estimates, DEMs, digital surface models, hydrological analysis, and aerial imagery.

RedCastle Resources has been providing geospatial helpdesk support to the Forest Service for several decades and has been promoting the appropriate use of remote sensing and geospatial technologies. Since 2004,we have provided helpdesk services to all units of the Forest Service through the Forest Service Customer Helpdesk (CHD). Services include support for geospatial software installation, licensing, mobile GIS and GPS applications, image processing, raster data modeling, various applications of remotely sensed imagery, and more. In addition staff assist users with acquiring server system accounts (e.g. ,AGOL, Collector, Field Maps) and troubleshoot issues with applications that may include issues with connections, data corruption, printing and display issues. This service provided to a large government agency, ensures users have knowledge, resources, and timely solutions for addressing a broad range of geospatial technical issues at all levels of the agency.

Riparian ecosystems are among the most diverse and important ecosystems but are also among the most threatened habitats in our everchanging world. Natural and anthropogenic changes, such as flooding, drought, wildfire, cattle grazing, and urban development can quickly alter a riparian corridor and threaten the fauna and wildlife that depend on it. For such high-risk sites, we use UAS technology to timely capture high-resolution thermal and natural color imagery for the use of monitoring and detecting changes. UAS thermal cameras allow us to detect cold and hotspots in rivers to assess for threats to trout habitat, and natural-color cameras provide the detail needed for mapping streambank erosion, riparian vegetation, rifle-pool systems, and presence of woody debris. This information enables us to provide real-time information to make informed decisions when it comes to supporting these dynamic ecosystems.

Mapping vegetation types is crucial for effective environmental management and ecological conservation. We have decades of experience creating robust vegetation maps covering extensive areas. We have conducted over 20 mid scale vegetation mapping covering diverse landscapes. Our in depth approach leverages a combination of remote sensing expertise, ecological background and advanced modeling techniques providing a comprehensive view of the landscape. This enables vegetation mapping that supports informed decision-making in forest and ecological management.

RedCastle Resources (RCR) partnered with the USFS to produce forest inventory models across the full extent of North Kaibab Ranger District (655,000 acres) bordering Grand Canyon National Park. Using a combination of Bayesian variable selection and linear regression modeling, we produced eight forest inventory models at a 20-meter resolution. In addition, we worked with pre- and post-fire lidar data to detect post-fire changes in vegetation cover and height. GTAC also produced a burn severity map based on canopy cover differencing and qualitatively compared it to existing fire severity products, including Monitoring Trends in Burn Severity and Rapid Assessment of Vegetation Condition after Wildfire data.

With over 300 pilots trained in mapping, our UAS mapping course covers essential topics such as camera settings, optimizing flight configurations, using flight planning software, conducting flight operations, ensuring accuracy, and processing data to deliver precise results—all while maintaining FAA compliance and safety. We also provide specialized training on advanced sensors, including thermal, multispectral, and lidar technologies. Our training leverages the expertise of our experienced pilots to provide your organization with the tools to conduct your own UAS work.

Understanding where, when, and what factors drive change across landscapes is essential for making informed land management decisions. To assist land managers and scientists, we helped develop the Landcover Change Monitoring System (LCMS) for the U.S. Forest Service through a collaboration between remote sensing experts from the USDA Forest Service, USGS, NASA, and several universities. LCMS is a remote sensing-based system that tracks landscape change across the United States from 1985 to the most recent full growing season. It provides the most accurate map of landscape change by utilizing advanced time series change detection techniques, Landsat and Sentinel data, cloud-based computing power, and big data analysis.

Highly detailed terrain and imagery provides critical information for safety and sustainability. We conducted a comprehensive aerial lidar and imagery survey covering 1,800 acres across 3 active mining sites and two geohazard sites near Bozeman, Montana and West Yellowstone, Wyoming. Utilizing a UAS-based lidar sensor, we captured high-resolution elevation data and imagery. This centimeter level accuracy dataset allows stakeholders to understand the terrain and vegetation to minimize environmental disturbances, identify potential hazards and implement mitigation measures to ensure safety. This data supports the effectiveness of remediation efforts and helps optimize long-term stability.