That challenge set the agenda for our theme at the 2026 Water Prediction Innovators Summer Institute: Advancing the Use of Quantitative Precipitation Inputs into the NextGen Framework. Ten graduate students in three teams (Don't Runoff!, FlashCast, and The Post-Processors) had seven weeks to bring sub-hourly radar rainfall, precipitation uncertainty, and ensemble forecasting methods into the NextGen Water Resources Modeling Framework. Seven weeks is not much, and the teams could not spend half of it compiling a modeling framework or downloading terabytes of radar data to laptops. Fortunately, they did not have to. The CIROH cyberinfrastructure team carried that load.

The framework fits in the Box​

NextGen is powerful but challenging to run. NextGen In A Box (NGIAB), CIROH's containerized distribution with more than 41,000 Docker pulls, turns that setup problem into a single pull command. This year's Summer Institute members were able to hit the ground running after just a single one-hour NGIAB workshop, demonstrating its sheer accessibility.

However, the teams did more than just run it. NGIAB is also an open box, and they took full advantage by extending it:

• FlashCast built a modified NGIAB Docker image that runs CFE and T-Route on a 15 minute timestep and forced it with 15 minute MRMS radar rainfall, expanding the framework's standard hourly configuration to the sub-hourly scales flash floods demand.
• The Post-Processors modified the NGIAB data preprocess library so NextGen could ingest MRMS radar estimates and HRRR forecast rainfall in place of the default AORC forcing. They then simulated 8,474 catchments around Ellicott City, Maryland.
• Don't Runoff! built a volume-preserving workflow that maps 2 minute, 1 km MRMS rainfall onto NextGen Hydrofabric catchments and used it to drive dHBV2-Flash, a new sub-hourly differentiable model designed for the framework.

Every modification lives in a container or a public repository. Any researcher can pull the same image, rerun the same experiment, and build on it. That is the point of the box: NGIAB enables maximal portability and reproducibility to maximize research's impact. For the complete set of shared resources, see each team's final report.

R2OHC Platform: NSF Jetstream2 and NGIAB did the heavy lifting​

None of this fits on a laptop. FlashCast alone ran 2,280 simulation scenarios per basin across three flash flood case studies (Asheville NC, Pittsburgh PA, and Kerrville TX) — 6,840 NextGen runs in total. The Post-Processors pushed 500 synthetic storms through NGIAB to train a neural network surrogate, on top of storm displacement experiments across thousands of catchments. Don't Runoff! processed radar rainfall for over 5,000 flash flood events across the country and trained a differentiable model on years of 15-minute data.

The computing resources came from CIROH cyberinfrastructure, which provided allocations on NSF Jetstream2, Indiana University's NSF-funded academic cloud, through CIROH's ACCESS allocation (EES240087). JetStream2 hosted virtual machines sized for each team's workload, from batch NGIAB runs to model training. The rainfall came from NOAA open data: 2-minute MRMS radar estimates, hourly AORC forcings, and HRRR forecasts, all pulled straight from public cloud archives. CUAHSI HydroShare closed the loop as the publication home for datasets and supplementary materials, and the NWC-CUAHSI-Summer-Institute GitHub organization hosts the code.

What the teams built​

Team Don't Runoff! delivered RUNOFF v1.0, a CONUS-scale catalog of over 5,000 flash flood events from the NOAA Storm Events Database, each mapped to hydrofabric catchments and paired with a USGS gauge and quality-controlled MRMS rainfall. RUNOFF is designed to conduct flash flood modeling experiments using high-quality precipitation estimates. They also delivered dHBV2-Flash, a 15-minute differentiable rainfall-runoff model that captured event peaks in the Upper Neuse basin with a median peak flow bias of 16% across 1,132 events. Supplementary materials are on HydroShare.

Team FlashCast addressed a question at the heart of flash flood prediction: how fine, and how accurate, does the precipitation input need to be? Their sensitivity pipeline converts thousands of NGIAB runs into critical uncertainty matrices, the precipitation uncertainty a forecast can tolerate before predictions stop being actionable. Early results from the three case studies point to a scale-dependent answer: for these events, resolution finer than about 7 km and 1 hour added little on its own, as long as storm volume was preserved. The pipeline is built to put that threshold to the test across many more basins and storms. Methods and figures are on HydroShare.

The Post-Processors built NCET, the Neighboring Catchment Ensemble Technique, which turns a single deterministic NextGen simulation into an ensemble streamflow forecast by borrowing hydrographs from neighboring catchments. Applied to the May 2018 Ellicott City flash flood with a real HRRR forecast, NCET recovered a flood the deterministic forecast missed, with the largest gains in probabilistic skill when the ensemble weights were aimed along the storm's displacement. That early result suggests storm location, more than catchment similarity, should steer the ensemble, and it points to a computationally inexpensive path toward flash flood forecasting within NextGen. The team also prototyped Replace and Route, a BMI component that lets NextGen borrow discharge from high-fidelity external models at hydrofabric nexus points. Evaluation materials are on HydroShare.

From asking research questions to prototyping workflows, in seven weeks​

Each product is hydrofabric-native, open, and packaged into containers, BMI modules, and reproducible pipelines — the way that the Community NextGen ecosystem expects. This ease of access and reuse is research to operations in miniature, and it is what CIROH cyberinfrastructure is for. NGIAB collapsed setup time from weeks to minutes. NSF's Jetstream2 scaled experiments beyond what any student laptop could run. Open data archives fed the models, and HydroShare and GitHub made every result citable and reusable. The students arrived with research questions. They left behind open prototypes, datasets, and code for the community to build on.

The 2026 Water Prediction Innovators Summer Institute took place June 8 — July 22, 2026. 24 graduate student fellows from 21 unique U.S. institutions formed research teams under the guidance of seven Theme Leaders, spanning topics from precipitation forecasting to flood risk communication. Research outcomes are summarized in the 2026 WPI SI final report.

The teams thank the CIROH Cyberinfrastructure & DevOps team and Hydroinformatics team at the Alabama Water Institute (Arpita Patel, Josh Cunningham, Quinn Lee, Nia Minor, and colleagues) for cloud provisioning and NextGen support throughout the program, Dr. Fred Ogden for guidance on developments with the NextGen framework, and the Summer Institute coordinators, Megan Vardaman and Nana Oye Djan. This work used Jetstream2 through allocation EES240087 from the NSF ACCESS program. This research was supported by the Cooperative Institute for Research to Operations in Hydrology (CIROH) under award NA22NWS4320003 from the NOAA Cooperative Institute Program.

At the University of Alabama's Surface Dynamics Modeling Lab (SDML), with support from CIROH in partnership with NOAA, we have been building the CIROH FIM Community Ecosystem, an interconnected, open-source, Python-based suite of toolsets for large-scale flood inundation modeling, simulation, benchmarking, and evaluation. Below we describe what each component of the Ecosystem does, how they work together, and how to get started.

The problem: fragmented capabilities without a unified FIM framework​

This effort starts with a simple question: can flood inundation modeling, benchmarking, and evaluation be brought together into a seamless, unified, and, most importantly, scalable system that serves both operational forecasters and the research community? What exists today are small-scale studies and individual tools: models predict floods, methods that evaluate accuracy, observations that record real events, each built for a specific task and demanding its own painstaking setup. What does not exist is a unified framework spanning modeling to evaluation at scale, for one study area or a thousand, so that methodologies are consistent and ready to calibrate models and test hypotheses. For operational forecasters, the missing piece is scale: mapping a single basin is a tractable problem, but operational forecasting requires generating and trusting a modeling approach across numerous river reaches in real time.

Evaluation raises the bar further. A model's true potential and limitations emerge only when it is tested at scale across many events, diverse floodplains, topographic settings, and hydrologic regimes. Large-scale evaluation is what turns a case study into a defensible conclusion about where a model works, where it fails, and why. Yet such evaluation depends on an "answer key": high-quality records of flood inundation for real flooding events. Those benchmark data have historically been scarce, scattered, and inconsistent, and where benchmarks are unavailable or of poor quality, evaluations lose credibility, and even skillful models are met with mistrust. The result: research groups rebuild the same pipelines from scratch, evaluations cannot be compared across regions or scales, and promising models take longer to reach the people who need them.

Our effort: the CIROH FIM Ecosystem​

Our answer is an interconnected, interoperable system of open-source components designed to work together through simple programmatic interfaces. Most components are deployed as standalone Python packages: install only what you need, or chain them into a complete workflow.

(a) FIMserv: FIM generation tool using the NOAA OWP HAND-FIM operational framework​

The Office of Water Prediction's Height Above Nearest Drainage (OWP HAND) framework, the operational FIM system of the United States, is a deliberately simple terrain-based model run over Hydrologic Unit Code 8 (HUC8) watersheds (more about HUCs) using National Water Model streamflow predictions across 2.7 million river reaches nationwide. Configuring it, however, is not straightforward: setting up the full pipeline from watershed data to streamflow and reproducing it across computing systems have been persistent obstacles to applying, improving, or studying an otherwise scalable model. FIMserv, the modeling component of the CIROH FIM Ecosystem, removes that barrier. Given a watershed of interest and an event, it automatically configures the entire modeling setup, retrieving the OWP-precalculated watershed data from CIROH's AWS S3 cloud storage (which mirrors the NOAA OWP-precalculated HUC8 datasets for the HAND FIM run) and the NWM streamflow for the chosen event (a past event, also called retrospective data; a live National Water Model forecast; USGS gauge records; or global hydrological products such as GeoGLOWS). It then leverages the OWP inundation mapping framework to produce a flood extent map within minutes (Baruah et al., 2025). It covers NOAA's full forecast range, from hourly short-range runs to 30-day outlooks, and supports simulation of many watersheds and flood events. It can also evaluate the NWM streamflow against USGS gauge observations, providing a measure of confidence in a map's inputs. Users can also modify key configuration inputs, such as river slope, calibration coefficients, and river bathymetry, supporting experimentation well beyond the default modeling setup. It is wrapped as a Python package that runs just as well on a laptop as on cloud infrastructure, so what once took a long time now takes a single installation command.

(b) FIMeval: FIM predictions evaluation framework​

A flood map is only as useful as the confidence placed in it, and that confidence comes from evaluation: comparing a model-predicted extent against a benchmark flood map from an actual or synthetic event. The traditional methods of evaluation, however, are manual, time-intensive, and error-prone; preparing maps in GIS platforms and/or doing it programmatically, and standardizing models and benchmarks (such as removing permanent water bodies and aligning projections) is so laborious that testing a model across many events becomes impractical. FIMeval automates, standardizes, and scales that step, evaluating any flood maps against benchmarks in a consistent, reproducible manner (Devi et al., 2026). It is deliberately model- and benchmark-agnostic: the map can come from any terrain-based, hydraulic, or data-driven model, and the benchmark from any trusted source. It computes several standard accuracy metrics, with built-in checks for how data imbalance between flooded and dry areas skews those scores. It then goes further, answering the impact-based question communities actually ask: which buildings does the model predict will be inundated, and was that prediction correct? It fits into any Python-based workflow: bring a flood map, whether produced by FIMserv or any other model, together with a benchmark you supply, and FIMeval scores the pair. What makes it especially powerful, though, is that users do not need to supply a benchmark at all: FIMeval connects seamlessly to the FIMbench database, so a single automated query pulls the right quality-controlled benchmark for your area and event and scores against it, provided the event date falls within our available benchmark coverage. FIMbench itself is our rigorously quality-controlled, multi-source benchmark database, described in detail in section (c) below. For FIMserv users, the evaluation can run embedded within FIMserv, generating and evaluating a map in a single pass. Demonstrations span major observed events and design floods alike across continental to global watersheds.

(c) FIMbench: an extensive FIM benchmarking database​

Trusting a model-predicted flood extent requires real-world evidence, precisely the benchmark data that has been hardest to obtain. FIMbench fills that gap with a rigorously quality-controlled library of multi-source, multi-sensor benchmark flood maps. Its current release includes 190+ maps (it will continuously keep growing with time) across the United States (Devi et al., in preparation), derived from aerial imagery acquired via NOAA's National Geodetic Survey (NGS), PlanetScope satellite imagery, Sentinel-1 radar, FEMA Base Level Engineering (BLE) 100- and 500-year return-period flow-based FIM products, and FIMs derived from USGS-surveyed high-water marks. Every map undergoes standardized processing and quality screening and is cataloged into quality- and product-based tiers; the guiding philosophy is "quality over quantity." FIMbench utilities then package each approved map into a database-compatible format and publish it to the CIROH AWS S3 database. Contributing to the database is currently limited to the SDML team, though we are exploring ways for the community to add data in the future; every published dataset, however, is openly available to everyone. Access follows two routes: the interactive FIMbench WebApp, developed with our partners at Aquaveo, supports visualizing, filtering, and downloading flood events; and the FIMbench Python API retrieves selected benchmarks programmatically from the cloud-hosted CIROH AWS S3 database. A cataloged, quality-assured, multi-sensor benchmark collection at this national scale has not, to our knowledge, been openly published before; paired with FIMeval, it makes testing a flood model against hundreds of real events practical rather than prohibitively laborious.

(d) FIMsens: flood extent and depth from satellite observations​

FIMsens, an upcoming Ecosystem component, adds an observation mode to the ecosystem: an end-to-end toolset that turns satellite imagery into error-corrected flood extent and depth products. Given a user-defined area of interest and event period, FIMsens discovers and previews suitable SAR or optical scenes, then processes them through linked modules for initial flood detection, reliability assessment, omission reconstruction, and depth estimation, beginning with Seeded Local Adaptive Thresholding (SLAT) (Wang et al., under review), which extracts the inundation directly observable in the imagery. The detections then pass to the Flood Object Reliability Filter (FOR-Filter) (Tian et al., in preparation), which uses image, geometric, terrain, and hydrologic evidence to remove or flag likely false positives before they propagate downstream. Finally, RS-FloodXDepth (Tian et al., 2026) combines retained flood evidence with surrounding topography to reconstruct hydrologically plausible inundation that would otherwise be missed due to vegetation, urban structures, clouds, or weak water signals, and to estimate depth from the inferred water surface and DEM elevation. A single pass delivers raw, filtered, and corrected extent layers together with a DEM-guided depth map.

Within the broader ecosystem, FIMsens complements rather than replaces the curated FIMbench reference databases. The same algorithm underpins FIMbench's Tier 2 and Tier 3 benchmarks, where its output is conservatively confirmed through qualitative assessment. Where the rigorously quality-controlled FIMbench database has a match, that curated map stays the trusted reference. Where no suitable benchmark exists, FIMsens can be used to generate event-specific, observation-derived flood extent and depth products from available satellite imagery, providing additional spatial and temporal coverage for FIMeval and related analyses. Because these event-specific products depend on the quality of the selected observations and processing conditions, they should still be reviewed through the toolset's QA outputs and, where appropriate, qualitative inspection before being used as reference information.

(e) FIMsim: cloud-based multi-model flood simulation​

FIMsim expands the ecosystem's flood-simulation engine, which models flooding using either terrain-based or hydraulic approaches: it runs the terrain-based NOAA OWP HAND-FIM through FIMserv, the ARC-Curve2Flood model, and extends the workflow to 2D hydraulic simulation using models with varying levels of physical complexity. Because no single flood model performs best everywhere, comparing alternatives has meant rebuilding a pipeline for each one; FIMsim removes that burden. In practice, the workflow is simple: the user defines a study area, and FIMsim acquires the inputs, builds the model configuration, and runs the hydrodynamic models of their choice, such as LISFLOOD-FP and TRITON. These hydrodynamic models provide GPU-enabled implementations that can substantially reduce execution time for computationally demanding simulations. Through the planned web application, users will be able to run the models remotely in a GPU-enabled Linux environment and download the results without installing the software locally. Different models can be run across catchments, or several models over a single catchment for agreement and ensemble analysis; integrating FIMeval and FIMbench provides evaluation for every output. Since FIMsim can be set up and run for every reach, a hypothesis can be tested across models rather than trusted to just one model. Simulation is not its only use. FIMsim can also serve purely as a preprocessor for hydraulic modeling, generating each input required for a 2D model separately. It can produce a mosaicked, clipped, reprojected, and resampled DEM at different spatial resolutions; generate Manning's n roughness grids from land-cover datasets of different sources, years, and resolutions; provide flowline data; and obtain inflow hydrographs from NWM, a USGS gauge, or an uploaded CSV file. The processed inputs can be exported in several commonly used formats, allowing modelers to use the ready-to-run files directly in their preferred hydraulic modeling engine. Today, FIMsim is available as a Python framework and a desktop application for macOS and Windows; a pip-installable Python library and a web application are also under development.

(f) FIMbox: a sandbox for large-scale, multi-model FIM development and analysis​

FIMbox is an open-source Python sandbox that treats flood inundation mapping as a controlled experiment: models, inputs, and parameters become variables that a researcher deliberately sets, holds fixed, or sweeps across in a single, reproducible run. A run begins with a boundary polygon for any area of interest rather than a precalculated HUC8, and a shared initialization step stages the common resources every model needs into one working directory: the DEM, NHD and NWM hydrography, FEMA NFHL flood hazard layers, NLD levees, OSM bridges and roads, and USGS gauge records. A configuration wrapper then exposes each subsequent stage, so a user-defined topography, hydrography, or roughness field is applied where it matters. The experiment setup determines which models to run, the terrain-based HAND workflow, a hydrodynamic model, or both. Each model triggers its own preprocessor, which builds only the datasets it needs and pulls inflows from the National Water Model or NextGen, or through the crosswalk module from global sources such as GEOGloWS.

Where FIMserv runs only the operational OWP HAND framework over precalculated HUC8 data, FIMbox opens the pipeline itself. Its HAND FIM modeling breaks the OWP HAND-FIM pre- and post-processing chain (from a DEM to a mapped extent) into small composable pieces: staging inputs, hydroconditioning the terrain, segmenting the stream network, generating HAND and its synthetic rating curves, calibrating them, generating the FIM, and evaluating it. Every stage exposes its parameters and writes intermediate products the user can inspect, replace, or feed back in, so a dataset or a coefficient can be substituted at one step and its consequences traced through to the mapped extent and depth. Because each dataset and parameter is a variable under the user's control, questions that are ordinarily laborious become routine: how much the resolution or the source of the river network, catchments, or DEM changes the predicted extent; how sensitive the result is to Manning's n roughness; and whether a better synthetic rating curve, a different stream segmentation, or an improved slope treatment sharpens extent and depth. Dask parallelization runs across the whole HAND pipeline, so the same workflow scales from a single basin to continental domains without rewriting a line of code, and the package runs the same way on a local workstation or on cloud computing resources.

What sets FIMbox apart is that a single experiment carries seamlessly from modeling through evaluation to benchmarking. Within one run it can drive a terrain-based HAND solution and, through the integrated FIMsim framework, a hydrodynamic model such as TRITON or LISFLOOD-FP, without altering their native configurations, then score each predicted extent with FIMeval against the best-matching FIMbench benchmark, selected by event date, geographic domain, spatial overlap, quality tier, or any combination of these; where no benchmark exists, FIMsens can attempt to derive one from satellite imagery. Because the whole chain, from fetching flows to scoring the result, is scriptable from a notebook or a scheduled job, FIMbox makes FIM research genuinely iterative: a hypothesis can be posed, tested, and scored at any scale, one basin or a thousand, with no manual handoffs between tools. That same design makes multi-model analysis natural, and work is now underway to extend the sandbox beyond a single model toward ensemble, multi-model FIM extents, repurposing results from a range of hydrologic models through the same workflow. Looking ahead, integrating NOAA's Next Generation Water Resources Modeling Framework (NextGen) will close the loop from meteorological forcing through hydrologic and inundation modeling to flood-extent assessment and evaluation, an end-to-end ecosystem simulation.

When should you use the FIM ecosystem?​

Each component can be used on its own; you do not need to adopt the entire ecosystem to benefit from a single component. However, integrating the components opens up research questions that no single component can answer alone. A few common starting points:

• "I need flood maps for a forecast or a past event or a synthetic scenario anywhere in the US; however, I need them fast and running on any local or cloud system." FIMserv can do that.
• "I built a flood inundation model, hydraulic, data-driven, or otherwise, and I have my own benchmark; I want to know how accurate the model is." FIMeval can do that.
• "I built a flood inundation model intended for regional, continental, or global application, and I want to investigate its potential and limitations across different hydroclimates and topographic settings." FIMeval, leveraging the FIMbench database, can do that.
• "I just need reliable benchmark flood maps for my study area." FIMbench, via the GUI or the Python API, can provide that.
• "I want to prepare the inputs for a 2D hydrodynamic model such as LISFLOOD-FP or TRITON, or run the model itself, wherever I need and without the usual setup hassle." FIMsim can do that, in a desktop app or Python framework, either handing you ready-to-run inputs and configuration to run on your own or running the simulation for you.
• "I want to run several flood models over the same watershed for ensemble and agreement analysis or assign a different model to each catchment according to what performs best there." FIMbox can do that, orchestrating the models through FIMsim and passing every result on to FIMeval and FIMbench for evaluation.
• "I want to experiment with the HAND workflow itself, swapping terrain, roughness, calibration, or model engines, feeding it custom or repurposed streamflow, and scaling from a single basin to continental domains." FIMbox can do that, breaking the workflow into flexible, Dask-parallel pieces that run straight through to evaluation.
• "I want to test a hypothesis on the terrain-based NOAA OWP HAND model, evaluating it across dozens or hundreds of events to judge whether the approach can be implemented across the US." The integrated ecosystem can enable that: FIMserv handles the modeling, and FIMeval, drawing benchmarks from FIMbench, handles the evaluation; the entire process is automated, programmatically and seamlessly.
• "I have no benchmark for my event and need to derive a flood extent and depth directly from satellite imagery." FIMsens can do that (once integrated), turning SAR or optical scenes into error-corrected extent and depth products.

Under the hood: technical dissemination​

Technically, the ecosystem is a modular, cloud-native suite of open-source Python packages and frameworks. What holds it together is a shared CIROH cloud backbone: every component reads and writes the same cloud-hosted data, so one tool's output is directly consumable by the next. FIMserv streams its inputs on demand: precalculated HAND hydrofabrics from the CIROH AWS S3 buckets, retrospective National Water Model (NWM) streamflow from NOAA's public S3 archive, and forecast NetCDF files from Google Cloud Storage, with the TEEHR framework's PySpark engine handling massive streamflow extractions. FIMeval builds on the standard geospatial Python stack (Rasterio, PyProj, GeoPandas, Shapely) for raster reprojection, map algebra, and dynamic bounding geometries based on different evaluation methods. The heavy layers it needs on the fly, such as permanent water bodies and Microsoft building footprints for impact-based assessment, are hosted on ArcGIS Online and other cloud services and fetched through REST endpoints for the evaluation domain only, so nothing oversized ever lands on the user's machine. FIMbench's dissemination rests on a decoupled web architecture: a Tethys Platform backend acts as a tile proxy, streaming compressed vector tiles of benchmark FIM extent straight from the CIROH AWS S3 storage to a React front end that renders interactive maps with MapLibre GL. The same S3 store serves the stateless FIMbench Python API, so the database remains a single, consistent source of truth, regardless of how it is accessed. FIMbox adds a parallel execution layer: it decomposes the HAND workflow into independent per-branch computations (hydroconditioning, branching, relative elevation, and synthetic rating curves) and uses Dask to distribute those branches across local cores or cloud workers. FIMsim moves computation to the cloud, provisioning the hydrodynamic engines on managed resources and assembling their inputs from public sources: USGS 3DEP terrain, land-cover-derived roughness, and NWM, USGS, or user-supplied hydrographs. Hardware is the one practical dividing line across the ecosystem: FIMbox's terrain-based HAND workflows run on ordinary local systems without a GPU, while the full FIMbox and FIMsim capability, terrain-based modeling coupled with 2D hydrodynamic simulation, is currently GPU-backed for accelerating the computation and Linux-based. FIMsens will add the observation path, query SAR and optical image catalogs, align downloaded scenes to a common DEM, and then hand them to the correction routine described above.

Who is it for? Try it today​

The CIROH FIM Community Ecosystem is completely open source and already in active use across the community: as of mid-2026, PyPI records put FIMserv past 45,000 downloads and FIMeval past 20,000, with the FIMbench API and the FIMbox testbed more recently released. How you start depends on the component: four of them are installed as Python packages with a single command (Astral uv makes it faster still), FIMsim ships as a desktop application, and FIMsens will join once released.

uv/pip install fimserve   # flood map generation

uv/pip install fimeval    # flood map evaluation

uv/pip install fimbench   # benchmark database access

uv/pip install fimbox     # FIM testbed access

Each framework is provided with step-by-step documentation and example notebooks on its GitHub repository: FIMserv, FIMeval, FIMbench, FIMbox and so on, so any user can go from installation to a first flood map and its evaluation without leaving the documentation. Run them wherever you work, from a laptop to Google Colab or CIROH's 2i2c JupyterHub. Prefer a point-and-click route? The FIMbench web portal lets you browse the benchmark database, and FIMsim runs today as a desktop application.

Who benefits from the FIM ecosystem? Broadly, anyone who works with flood hazard modeling. From researchers testing new models to forecasters, planners, engineers, and students, each takes something different from it: a place to experiment, options they can defend in operations, and flood information that has been checked before it is trusted. Our goal is simple: anyone, whatever their background, can make a flood map, check it against reality, and trust what it tells them.

Get involved​

The FIM ecosystem is a living effort, and feedback will improve it. We will appreciate your feedback and suggestions about technical problems, proposed utilities, and additional components. We are committed to incorporating that input into each release, and collaboration is what we need most. The ecosystem is still very much in motion: not every framework yet has its own graphical interface, and some are not yet packaged for PyPI, but many avenues and dimensions are advancing in parallel. Keep an eye out for new releases, and email us about collaborations, ideas, or anything you would like to build with us. All the FIM components code lives in the SDML GitHub repository, and each repository has its own contribution guidelines: open an issue with a new idea or a bug, submit a contribution, or simply reach out to the SDML team. We would genuinely love to hear from you; it is how this ecosystem will keep evolving. And this is only a start: the CIROH FIM Community Ecosystem is not limited to the components described here. We want it to grow beyond SDML, with colleagues across CIROH institutions and the broader community contributing new components, datasets, and ideas that strengthen it as a shared, community-owned capability.

Acknowledgments​

This work is supported by the National Oceanic and Atmospheric Administration (NOAA) through the Cooperative Institute for Research to Operations in Hydrology (CIROH) under the NOAA Cooperative Agreement with The University of Alabama (NA22NWS4320003). We thank the Aquaveo team for collaborating on the development of the Graphical User Interfaces for these tools.

CIROH cyberinfrastructure used: CIROH AWS S3 cloud storage hosts the OWP-precalculated HUC8 HAND hydrofabric datasets streamed on demand by FIMserv, and hosts the FIMbench benchmark database; the CIROH Tethys Platform / CIROH Portal hosts the FIMbench WebApp, developed with our partners at Aquaveo (we are also thinking of hosting a curated virtual environment/image on 2i2c services soon). This cyberinfrastructure is the backbone that makes the ecosystem scalable and interoperable: because every component reads from and writes to the same CIROH cloud-hosted data, one tool's output is directly consumable by the next, and users never need to download oversized national datasets. CIROH's Tethys hosting and 2i2c JupyterHub lower the barrier further, providing point-and-click and zero-install routes into the tools, while CIROH compute resources make evaluation across hundreds of events and continental-scale runs practical rather than prohibitive.

Additional Links​

• Understand the CIROH FIM ecosystem visually: https://sdml.ua.edu/fim-ecosystem/
• Explore the CIROH FIM ecosystem more interactively with our NotebookLM Mind Map

Publications​

1. Baruah, A., Dhital, S., Cohen, S., Tran, T.N.D., Elhaddad, H., Watts, C.L., Devi, D., Chen, Y., & Pruitt, C. (2025). FIMserv v.1.0: A tool for streamlining Flood Inundation Mapping (FIM) using the United States operational hydrological forecasting framework. Environmental Modelling & Software, 192, 106581. https://doi.org/10.1016/j.envsoft.2025.106581
2. Devi, D., Dhital, S., Munasinghe, D., Cohen, S., Baruah, A., Chen, Y., Tian, D., & Pruitt, C. (2026). A framework for the evaluation of flood inundation predictions over extensive benchmark databases. Environmental Modelling & Software, 196, 106786. https://doi.org/10.1016/j.envsoft.2025.106786
3. Devi, D., Munasinghe, D., Tian, D., Dhital, S., Cohen, S., Raghavan, R., Swain, N., Nikrou, P., Baruah, A., & Chen, Y. (in preparation). FIMbench: A benchmark database of multi-tier flood inundation maps for large-scale evaluation.
4. Tian, D., Liu, H., Wang, L., Cohen, S., & Mandal, T. (2026). RS-FloodXDepth: Enhancing remote sensing-derived flood extent and estimating flood depth using a hydrologically guided region-growing method and high-resolution DEMs. Water Resources Research, 62, e2025WR042384. https://doi.org/10.1029/2025WR042384
5. Tian, D., Liu, H., Cohen, S., & Wang, L. (in preparation). Commission error analysis and correction in remote sensing flood maps using an object-oriented, hydrologically informed approach.
6. Wang, L., Liu, H., Tian, D., & Cohen, S. (2025, under review). Rapid automated flood mapping using SAR imagery and locally adaptive thresholds derived from multivariate gamma probability distributions. International Journal of Applied Earth Observation and Geoinformation.

Our Team​

Supath Dhital

Geography Researcher II

The University of Alabama

Dipsikha Devi

Postdoctoral Researcher

The University of Alabama

Anupal Baruah

Research Scientist

The University of Alabama

Dinuke Munasinghe

Research Scientist

The University of Alabama

Parvaneh Nikrou

PhD Candidate

The University of Alabama

Dan Tian

Postdoctoral Researcher

The University of Alabama

Yixian Chen

Research Scientist

The University of Alabama

Sagy Cohen (PI)

Professor

The University of Alabama

What we provisioned, by the numbers​

Running an event of this scale requires more than just a single cloud context, as each workshop posed its own computing and data access requirements. CIROH Cyberinfrastructure met this challenge via a coordinated stack spanning multiple cloud providers and services. Across the 15 workshops that needed infrastructure (14 requests), the R2OHC cloud provisioned the following:

One link, no setup: the Workshop Hub​

For 12 of these workshops, their backbone was the CIROH–2i2c JupyterHub, a managed JupyterHub environment that 2i2c operates on Google Cloud. For the conference, we provisioned a dedicated Workshop Hub: an ephemeral environment that lives only for the duration of the event and then spins back down.

The work that happens before anyone logs in: custom images​

Nine of the twelve JupyterHub workshops needed a custom software image to support their unique computing needs. Sessions covering topics like the Community Asset Computation Hub (CCNH), Foundations of Machine Learning, CUAHSI Hydroinformatics, HydroServer, INFLECT-based Flood Inundation Mapping, Satellite Data Projection, Machine-Learned Model Emulators, and Water Quality Modeling from Hillslope to Watershed Scales each had distinct dependency stacks.

That's a coordination problem, and it was solved weeks before the conference. Workshop leads told us what their sessions needed, we defined those environments in our public awi-ciroh-image repository on GitHub, and 2i2c deployed them. By the time attendees arrived, the right libraries were already in place, all but eliminating the usual loop of haggling with installations and configurations that “work on my machine”. Instead, there were reproducible environments, ready and consistent on day one, demonstrating clear evidence of how tightly the CIROH and 2i2c teams work together.

CCNH: Instant Access to NextGen in JupyterHub​

Two of the twelve workshops also marked the first appearance of the CIROH Community NextGen Hub (CCNH) at a DevCon event! This JupyterHub image contains all of the dependencies required to run the NextGen Framework, bypassing the framework's notorious complexity. CCNH represents a new paradigm where the NextGen framework can be run entirely from the comfort of a web browser, further extending its accessibility and expanding the potential for research and development with the framework.

DevCon 2026 workshops using CIROH Cyberinfrastructure​

• #463 — Hands-on LSTM and Transformer for Operational Streamflow Prediction (Penn State): Brings state-of-the-art deep-learning sequence models into an operational streamflow forecasting workflow. This workshop made use of 2i2c GPU servers to enable its cutting-edge approach.
• #462 — Foundations of ML: From Architecture to Optimization (UA): Entry-level on-ramp giving hydrology researchers the core building blocks of machine learning workflow.
• #460 — Machine Learning: Satellite Data Projection and Georeferencing with Satpy (Stevens): Teaches the remote-sensing preprocessing (projection/georeferencing of polar-orbiting data) needed to feed satellite inputs into models.
• #458 — Essential Geospatial Data and Coding Skills for CIROH Researchers / CUAHSI Hydroinformatics (CUAHSI): Establishes the baseline geospatial-data and coding literacy that underpins nearly every other track.
• #455 — Physics-Based Framework to Predict Water Quality from Hillslope to Watershed Scales (U. Iowa): Applies a physics-first modeling approach to demonstrate the design decisions that go into scaling a model from local to regional contexts.
• #449 — Topographically Derived High Flow Thresholds for Flood Inundation Mapping (USU): Demonstrates an approach to improve flood-inundation mapping by deriving high-flow thresholds directly from terrain.
• #441 — Temporary API Key for CIROH NWM API: Two workshops queried large hydrologic datasets directly through the Google BigQuery API, one in Flood Inundation Mapping and one in Hydroinformatics.
• #411 — Talk to NRDS: LLM Chat for TethysDash to Explore the NextGen Research Data Stream (BYU/Aquaveo): Demonstrates a natural-language/LLM interface for querying NRDS outputs, lowering the barrier to exploring the data stream. (Notably, this workshop used 50 custom-provisioned VMs on the JetStream2 cloud to handle the heightened GPU requirements of this novel technology.)
• #410 — Machine-Learned Emulator Models for Water-Quality Uncertainty (UVM): Trains ML emulators on the output of baseline models to quantify predictive uncertainty cheaply, replacing expensive repeated runs of in-depth physical simulations.
• #408 — Integrating Models in NGIAB, NRDS (UA): The core research-to-operations session covering the model-integration lifecycle from BMI into NGIAB and onward to NRDS.
• #389 — Managing and Standards-Based Sharing of Hydrologic Observations Using HydroServer (USU): Standardizes how observational data is collected, organized, and shared via built-for-purpose software, solidifying the evaluation/ground-truth side of modeling.
• #388 — Orchestrating End-to-End Reproducible NextGen Workflows (U. Calgary): Focuses on reproducibility, chaining NextGen and adjacent steps into repeatable, shareable workflows. This session made use of both a single "Community Asset" image and a shared storage volume, with input data pre-staged in a common folder so dozens of users could work against the same datasets across both sessions.
• #383 — Uncertainty Analysis for Extreme Event Hydrometry (U. Iowa): Provides the theory and tools for quantifying uncertainty specifically in extreme-event measurement.
• #379 — Interactive CIROH Research Apps with Python-only Tethys Component Apps (Aquaveo/BYU): Lets researchers build interactive web apps for their results in pure Python, without the overhead of learning unwieldy front-end frameworks.

HydroShare integration: from a resource to a running notebook​

Many workshop materials were published as HydroShare resources and launched straight into the Workshop Hub via a single link, pulling notebooks and data into the user's environment automatically. What might otherwise have been an onerous process of managing cluttered Downloads folders and local directory structures was instead compressed into a single click.

Why the 2i2c partnership matters​

2i2c's significance in this process went far beyond just hosting the JupyterHub. 2i2c specializes in open-source cloud infrastructure for research, and that shared commitment shaped our collaboration. Our entire stack — JupyterHub, the container images, the HydroShare connector — is open source and lives in public repositories. Another institution could easily read our Dockerfiles or dependency configurations to launch similar environments.

The operational model matters too. 2i2c manages JupyterHubs for many science communities, and what they learn from one community gets shared quickly with others. Our pre-conference testing and prep was informed by dashboards 2i2c built for EarthScope's computing-intense workshops, and the temporary storage approach 2i2c and CIROH used at DevCon is now being re-used with other communities (including EarthScope).

Lessons we're carrying forward​

Pre-event automation is everything. Nine custom images don't build themselves on the morning of a workshop. Front-loading that work meant we could focus on people, not infrastructure, during the event.

Treat infrastructure as tracked engineering. Thirteen GitHub issues turned a sprawling set of requests into an auditable plan — and a head start on next year.

Reproducibility scales. Version-controlled environment images are the difference between 15 consistent workshops and 15 individual troubleshooting sessions — and they let any of these workshops be re-run on the mainline (NM2.1) 2i2c JupyterHub environment.

Open-source, public-private partnerships work. Corporate cloud providers (AWS/Google Cloud), an NSF-funded computing resource, a mission-driven infrastructure partner with 2i2c, and a university research institute (CIROH) all contributed aspects of the DevCon experience. The result was a unified experience across every workshop. That's the model we want to keep building on.

To our partners at 2i2c, and to the providers and resources that backed DevCon 2026: thank you. To our DevCon networking sponsors — Lynker, Google Cloud, Civil and Environmental Engineering at The University of Utah — thank you.

And to everyone who logged in, ran the notebooks, and pushed their work forward — that's exactly what this infrastructure is for.

Interested in CIROH's cloud resources? Learn more at hub.ciroh.org/docs/services/intro or reach the CIROH Cyberinfrastructure team at ciroh-it-support@ua.edu.

Kriging-Based Streamflow Estimation​

Process-based hydrologic models are subject to structural and forcing uncertainties throughout the modeling domain, yet these can only be evaluated where USGS gauge observations exist. There is a clear need for a data-driven, observation-based framework that provides spatially complete streamflow estimates with well-characterized uncertainty, independent of model structure. Recent results from the CIROH project "Developing and Benchmarking Data Assimilation Methods on a Standardized Testbed" suggest that a simple Kriging interpolation between USGS gauged locations is both scalable and accurate for producing such spatially complete streamflow fields. As a pure data-driven method, this interpolation cannot be used directly for forecasting, but it serves as a valuable "pseudo-observation" for streamflow analysis and historical reconstruction.

Hourly Streamflow Kriging Across CONUS​

Streamflow kriging effectively leverages spatial correlations in observations and has previously been implemented in the National Hydrologic Model for calibration and evaluation at daily timescales (Farmer et al., 2019). Here, we extend this approach to hourly resolution using instantaneous-value (IV) discharge records from approximately 7,300 active USGS gauges across CONUS. Instantaneous values are aggregated to hourly means, and variogram parameters are estimated using a spherical variogram model. These parameters drive gridded hourly interpolated runoff and kriging error variance fields across CONUS, which are then used to estimate distributed runoff at every catchment divide within the NextGen hydrofabric.

Integration into the NextGen Research Data Stream​

To make hourly kriging operationally available, we integrated it into the NextGen Research Data Stream (NRDS; ngen-datastream#352). Kunal Sarna designed the software stack aligning the outputs in a long-format Brotli-compressed Parquet schema that downstream consumers such as T-Route expect, and built a Docker image packaging the kriging workflow (docker.io/awiciroh/qkrig:2.2.0). Harsha Vemula led and executed all aspects of deployment: configuring the AWS infrastructure, setting up the EventBridge schedule, provisioning and managing the nightly EC2 instance, orchestrating the end-to-end data pipeline and delivery of outputs to the NRDS status dashboard.

The NRDS at CIROH-UA now runs qkrig as a first-class datastream, updating the previous 24 hourly estimates once per day. Each night at 00:30 ET, the pipeline:

1. Fetches USGS instantaneous-value discharge for ~7,300 active gauges
2. Fits an ordinary kriging field across CONUS at hourly resolution
3. Samples the field at every NextGen v2.2 catchment centroid

The full daily output — 24 hourly kriged NetCDFs and variograms, kriging plots and a daily GIF, and a single ~13 MB Parquet file covering all 831,777 catchments × 24 hours of streamflow — is published publicly at:

s3://ciroh-community-ngen-datastream/outputs/qkrig/qkrig.{YYYYMMDD}/

and tracked on the NRDS status dashboard alongside other operational datastreams.

Figure 1: Hourly kriged streamflow and variogram for the CONUS region, accessed from the NextGen Research Data Stream: https://datastream.ciroh.org/#outputs/qkrig/qkrig.20260519/plots/kriging/

Streamflow Simulation Using T-Route​

To evaluate the utility of kriging for generating spatially distributed runoff in ungauged basins, we interpolated runoff over the South Toe River hydrofabric by withholding observations from 20% of USGS gauges affected by Hurricane Helene in September 2024. The resulting kriged runoff was used to initialize T-Route, a river-routing engine developed by NOAA's Office of Water Prediction. Sonali Vyas implemented streamflow simulation with interpolated runoff at hydrofabric catchments using T-Route. Comparison of routed streamflow at the gauge outlet with observations showed that the kriging-initialized T-Route simulation captured both the peak timing and recession during Hurricane Helene. This demonstrates that combining streamflow kriging with T-Route is a viable approach for streamflow simulation (with lag times of 1–24 hours) in ungauged basins within the NextGen water resources modeling framework.

Figure 2: T-route simulation of streamflow initialized using held-out kriged runoff in the NextGen hydrofabric of the South Toe River.

Summary​

Hourly streamflow kriging is now operational within the NextGen Research Data Stream, delivering spatially complete estimates for all NextGen v2.2 hydrofabric catchments. This observation-based approach supports a broad range of potential applications including streamflow analysis and historical reconstruction, National Water Model calibration, streamflow forecasting, and data assimilation for ungauged basins.

Links​

1. GitHub repository — streamflow kriging (qkrig)
2. GitHub repository — T-route streamflow simulation with kriging
3. Streamflow analysis from kriging DataStream
4. Daily streamflow reanalysis dataset based on kriging
5. Hourly streamflow reanalysis dataset based on kriging

Acknowledgements​

Special thanks to Kunal Sarna and Harsha Vemula for collaborating and integrating hourly kriging into NRDS in 3 weeks.

This research was supported by the Cooperative Institute for Research to Operations in Hydrology (CIROH) through NOAA Cooperative Agreement NA22NWS4320003. CIROH cyberinfrastructure used: AWS S3 (daily and hourly streamflow reanalysis products from 1979–2025 WYs), EC2 (testing the kriging workflow in NRDS), and DataStream (hourly streamflow analysis outputs).

Why do we need TEEHR?​

TEEHR is built on a fundamental question: “Which hydrologic model is better?”. At its simplest, this can seem trivial; the model simulation is paired with observations, a few performance metrics are calculated, and you're on your way to a common performance analysis. Things can start to get more complicated if want to go larger, dig deeper or ask more nuanced questions:

• What if we want to compare many models against each other?
• What if we want to analyze thousands of locations with 40-years of hourly timestep data at the continental scale?
• What if we want to interrogate the data with questions like:
  • “How does performance during high-flow events compare to low-flow events?”
  • “How does model performance relate to physical basin attributes?”
  • “What's the uncertainty associated with the resulting metrics?”
• What if we want to make the data easily accessible to the hydrologic community to support both historical and near-real time analyses?

These are the challenges TEEHR is designed to address.

What is TEEHR?​

The TEEHR Framework​

TEEHR began as a Python package enabling robust and iterative evaluation of hydrologic models. It has since expanded into a cloud-based platform (TEEHR-Cloud) consisting of a high-performance data warehouse optimized for huge analytic tables, a suite of cloud-based services supporting data exploration, automated processing and data ingests, and web-hosted interactive dashboards.

TEEHR-Cloud​

TEEHR-Cloud delivers evaluation-ready insights and datasets that are publicly available and continuously updated — including 40 years of hourly retrospective streamflow simulations across more than 8,000 gage sites, model forcings summarized to drainage basins (mean areal precipitation and temperature), and a real-time evaluation manager for live data ingestion and analysis of operational NWM, RFC forecasts and CIROH’s NextGen Research Data Stream (NRDS). It's all built on Apache Iceberg, an open table format designed for the demands of big data at scale and proven by well-known big data industries (e.g., Netflix).

TEEHR-Python​

TEEHR-Python is the analytics engine that puts this data to work. Powered by PySpark for scalable, distributed compute and Apache Iceberg for robust data management — including time travel, upsert/append/delete operations, schema evolution, and ACID transactions — it gives analysts the full power of modern data engineering in a hydrologic context. An ever-growing suite of signature, deterministic, and probabilistic metrics, with sampling uncertainty estimation built-in, means users can interrogate model performance in ways that simply weren't easily accessible before. TEEHR-Python is still a Python package that users can install in their own environment via pip install teehr to use with their own data and to access datasets in the TEEHR-Cloud data warehouse.

What are the TEEHR-Cloud Dashboards?​

Several web-based dashboards, currently in development, allow users to view and interact with the data warehouse from application-specific perspectives. The two active dashboards include more general retrospective model comparisons and forecast analysis, while other, more application-specific, dashboards in development will focus on:

• Operational forecast (NWM, RFC) diagnostic and event-based evaluation
• Multi-model intercomparison to baselines with uncertainty
• Ensemble water supply forecasts
• Forecast Informed Reservoir Operations (FIRO)
• Flash flood forecasts
• Low flow forecasts

Summary​

The TEEHR framework is under active development, as we continue to expand historical and real-time data ingests, application-specific dashboards, and core TEEHR-Python functionality. To learn more about TEEHR, the documentation is a good place to start (link below). If you want to ask a question, request a change or new feature, or report a bug, you can reach out through the CIROH #teehr slack channel, by creating an issue on the TEEHR GitHub repos, or by contacting us directly. Thanks for reading!

Additional Links:​

• Documentation: https://hub.ciroh.org/docs/products/evaluation/rtiteehr/
• TEEHR documentation: https://rtiinternational.github.io/teehr/
• Dashboards (in-development): https://dashboards.teehr.rtiamanzi.org/
• TEEHR repo: https://github.com/RTIInternational/teehr
• TEEHR-Cloud repo: https://github.com/RTIInternational/teehr-hub

Our Team​

Matthew Denno (PI)

Lead Software Developer

RTI International

mdenno@rti.org

Katie van Werkhoven

Lead Science Advisor

RTI International

kvanwerkhoven@rti.org

Sam Lamont

Lead Software Developer

RTI International

slamont@rti.org

Sam Landsteiner

Software Developer

RTI International

slandsteiner@rti.org

The NextGen Research Datastream (NRDS) provides a mechanism by which these ideas can be refined and make their way into operations.

Developed by Lynker and the Alabama Water Institute (a Cooperative Institute for Research to Operations in Hydrology partnership), the NRDS facilitates the actualization a research idea from the community in a scalable and deployable numerical prediction system. To evaluate each of these modeling concepts, NRDS deploys prototype models to generate a continuous “datastream”. These outputs can then be evaluated and made more accurate. This cycle of streamlined deployment and iterative design lets these prototypes mature into a product that can be picked up by an operational forecasting team.

To enable this process to be done rapidly and smoothly, the entire system is designed with reproducibility and iterative improvement as core principles. The NRDS is an automated numerical prediction system generating regular stream flow forecasts that uses the NextGen Water Resources Modeling Framework (NextGen) as the core modeling engine and NextGen In A Box (NGIAB) as the simulation environment. This system generates forecasts across the contiguous United States (CONUS) on CIROH's operational cyberinfrastructure backbone: the research-to-operations (R2O) Hybrid Cloud (R2OHC) platform, with deployment on the AWS cloud. What makes the NRDS exciting is that the entire system is open-sourced, reproducible, publicly browsable, and potentially editable by anyone in the hydrologic community.

Why the NRDS Matters​

For decades, operational hydrologic modeling has been a largely closed process. National-scale models like the National Water Model (NWM) are developed and maintained by federal agencies where HPC access is limited. While these models frequently produce insights and forecasts of excellent quality, the simulations themselves have proven difficult to reproduce and improve upon from within the hydrologic research community, hindering the knowledge flow from research to operations. By contrast, the NRDS is totally public and open. This provides both researchers and national agencies with immediate access, reducing the latency between the discovery of new and potentially valuable knowledge and its operational implementation.

Deployed on AWS cloud from the ngen-datastream repository, the NRDS regularly ingests meteorological forcing data, processes it through the NextGen framework using community-configured parameters, and produces streamflow predictions for river reaches across the country. The forcing data, model configuration, infrastructure, and simulation outputs are all publicly available through the NRDS S3 data portal and the NRDS visualizer. Researchers can browse the results, compare them against observations, and reproduce or run their own simulations with the exact tooling that supports the NRDS. This transparency and usability enables researchers to readily propose improvements to any component of the NRDS system stack.

This is a fundamentally new paradigm for how national-scale water prediction can work: an open, iterative, community-driven feedback loop where regional expertise flows directly into an operational-like system.

How the Community Can Contribute​

This open contribution model is perhaps the most exciting aspect of the NRDS. The configuration files and NextGen model formulation that drive the daily simulations are publicly hosted and directly referenced during every execution. This means that community-proposed improvements to parameterization can flow directly into the operational system just by modifying these files.

The process is intentionally straightforward. A researcher who has improved upon the parameterization of a particular VPU realization file can navigate to the ngen-datastream Issues page on GitHub, open a new issue using the dedicated "Propose NRDS/NextGen Research DataStream parameter update" template, and submit their updated configuration along with supporting evidence. The CIROH team then reviews these contributions; once validated, they are merged and deployed into the live NRDS system.

This contribution pathway was a central theme at the 2025 CIROH Developers Conference (DevCon 2025), where a dedicated workshop walked participants through the complete process of proposing a configuration change. Attendees received access to virtual machines on NSF JetStream2 with pre-installed tooling, enabling them to run local simulations, compare results, and prepare a contribution — all within a single workshop session.

The NRDS is not designed to be a product delivered by a central team. It is a platform designed to absorb and amplify the collective expertise of the hydrologic community.

Multiple Datastreams Live and Scaling​

The long-term potential of the NRDS is further amplified by its scalability in deploying additional model formulations to generate regular forecasts. The system now runs multiple concurrent datastreams, each using a different hydrologic modeling approach within the NextGen framework to generate streamflow forecasts.

Currently, the NRDS operates a CFE-NOM datastream across all CONUS and parallelized across Vector Processing Units (VPUs), which pairs the Conceptual Functional Equivalent (CFE) rainfall-runoff model with the Noah-OWP-Modular land surface model. Alongside this, an LSTM DataStream has been deployed for all CONUS, bringing cutting-edge machine learning-based streamflow prediction into an operational-like system. Both of these datastreams route the NextGen outputs using T-Route.

The most recent release, v2.2.0 (February 2026), deployed the LSTM model into the production NRDS system — a milestone that brings differentiable and machine learning-based hydrology into an automated, cloud-based production environment. This builds on the broader CIROH effort to integrate models like δHBV 2.0 from Penn State's MHPI group.

There is also a Routing-Only DataStream for VPU 03W. This datastream focuses on channel routing NWM outputs using T-Route only, a novel experiment by Quinn Lee.

The Broader Ecosystem​

The NRDS is deeply interconnected with the broader suite of CIROH tools that are making NextGen modeling accessible. NextGen In A Box (NGIAB) provides the containerized runtime environment that researchers can use locally to test configurations before proposing them to the NRDS. The DataStreamCLI (now in its own repository) automates the complete workflow from data preprocessing to NextGen execution. The ForcingProcessor (also split out) handles the conversion of gridded meteorological data into the catchment-averaged forcings that NextGen requires. Once the model outputs are generated, TEEHR (Tools for Exploratory Evaluation in Hydrologic Research) provides standardized evaluation capabilities for comparing model outputs against observations (dashboard available here).

Together, these tools form a coherent pipeline: prepare data with the ForcingProcessor and DataStreamCLI, run simulations locally with NGIAB, evaluate results with TEEHR, and — if improvements are found — contribute them back to the NRDS for operational deployment.

What's Next​

The trajectory of the NRDS points toward an increasingly capable and diverse operational system. Active development efforts include implementing the hourly, deep learning-based differentiable model δHBV 2.0 MTS, which was recently embedded into the NGIAB-NRDS ecosystem through a collaboration between Penn State and the Alabama Water Institute. This model has demonstrated continental-scale streamflow forecasting capabilities that rival the current National Water Model.

There is also ongoing work to make the NRDS outputs more accessible and useful to downstream consumers. The NRDS Visualizer now provides a web-based interface for browsing and exploring daily simulation results. The TEEHR Evaluation Dashboard now provides rapid evaluation of these simulation results, enabling researchers to gauge model performance as events are occurring and weigh the performance of each datastream (NextGen model-formulation) relative to other datastreams.

With NRDS ecosystem quickly reaching maturity, now is the time for hydrologic researchers to put their research ideas to the test and deploy them in this operational-like environment.

Get Involved​

The NextGen Research DataStream represents a new kind of infrastructure for water science: one that is as much a social and organizational innovation as it is a technical one. By making national-scale hydrologic simulations open, reproducible, and community-editable, the NRDS creates the conditions for a distributed network of research hydrologists to incrementally improve the accuracy and reliability of streamflow predictions across the country.

Whether you are a graduate student exploring NextGen for the first time, a river forecast center hydrologist with deep regional expertise, or a researcher developing novel modeling approaches, the NRDS offers a concrete pathway from your work to operational impact.

Here's how to get started:

• NRDS on the NextGen in a Box product portfolio website: ngiab.ciroh.org/#/nrds
• Explore the data: Browse daily outputs at datastream.ciroh.org or the NRDS Visualizer
• Examine the NRDS deployment status timeline to see which datastreams have been deployed, when, and at what scale: NRDS Status
• Propose an improvement: Use the NRDS Issues page to submit your idea for a new datastream or an edit to an existing deployment.
• Join the conversation: Participate in GitHub Discussions to connect with the community
• Read the documentation: Visit the CIROH Hub NRDS page for comprehensive documentation on the NRDS system and broader ecosystem.
• Try out NRDS tools: Clone the various NRDS repositories to experiment with the NRDS workflow.
  • DataStreamCLI (Figure 2): the on-server workflow tool used in NRDS simulations. This tool offers the ability to reproduce NRDS simulations, with a modular design for integrating research configurations and processing components.
  • ForcingProcessor (Figure 3): a scalable tool for processing NWM operational data files into NextGen inputs. This tool provides the processing to generate NRDS forcings.

The future of water prediction is open. Come build it with us.​

The NextGen Research DataStream is developed and maintained by CIROH at the University of Alabama, with funding under award NA22NWS4320003 from the NOAA Cooperative Institute Program. Learn more at ciroh.ua.edu.

The NextGen framework holds great potential for hydrologic modeling, but is often inaccessible due to its strenuous setup and requirements. As such, embedding it within a cloud-based framework offers a natural solution to this problem by removing some of the administrative and technical requirements for compute resource setup and computational library configuration, thus opening the door for a wider audience to tke advantage of the strengths of the framework.

With the CIROH Community NextGen Hub (CCNH), we’ve created a cloud-based environment that addresses exactly those setup challenges, so users can focus on science instead of software.

A Preconfigured, Ready-to-Use Cloud Environment​

CCNH is a containerized, cloud-based modeling environment hosted on the CIROH-2i2c JupyterHub. It packages everything a researcher needs to run end-to-end NextGen workflows — from input preprocessing through model execution, calibration, evaluation, and output visualization — into a single, ready-to-use JupyterHub image. Built on the same containerization patterns as NGIAB, CCNH leverages a Pangeo base image and includes:

• Pre-compiled NextGen framework binaries from NGIAB based docker image
• NGIAB data preprocessing tools for automated retrieval and subsetting of hydrofabric and meteorological forcing datasets
• T-Route routing components for streamflow simulation
• SPOTPY(Statistical Parameter Optimization Tool for Python) for model calibration
• TEEHR(Tools for Exploratory Evaluation in Hydrologic Research) for performance evaluation
• PyNGIAB, a Python wrapper that lets you run NextGen simulations directly from Jupyter notebooks
• HydroShare integration tools (nbfetch, hs_files-jupyter, hsclient) for seamless data exchange to save results in HydroShare for collaboration, reproducibility and publishing
• JupyterLab with distributed computing capabilities for interactive, scalable workflows

The result: researchers can go from zero to running a calibrated NextGen simulation in a fraction of the time previously required.

Streamlined Workflow, Clear Results​

Through guided notebooks, users can:

• Prepare and subset input data
• Run simulations
• Calibrate models
• Evaluate performance
• Explore outputs interactively

...all within a single environment designed for efficiency and discovery.

Why It Matters​

By lowering technical barriers and using a cloud-based approach, CCNH makes NextGen accessible to a broader research community. It allows scientists to spend less time on setup and more time advancing hydrologic understanding.

How to Access CCNH​

There are two primary ways to launch CCNH in the CIROH-2i2c JupyterHub environment:

Option 1: Launch Directly from CIROH-2i2c JupyterHub​

1. Go to: https://ciroh.awi.2i2c.cloud/
2. Click “Login to continue” and authenticate with your CIROH credentials. (If this is your first time, you may be prompted to authorize access through GitHub.)
3. On the Server Options page:
   • Select the desired compute profile (Small, Medium, Large, or Huge).
   • Select the CIROH Community NextGen Hub image.
4. Click Start to launch the JupyterHub.

You are now inside a CIROH 2i2c JupyterHub session running the CCNH image, which provides a fully configured NextGen modeling environment.

Option 2: Launch from a HydroShare Resource (“Open With”)​

One of the most powerful features of CIROH-2i2c JupyterHub is its integration with HydroShare. You can open a HydroShare resource directly in a CIROH-2i2c JupyterHub session running the CCNH image:

1. Navigate to a HydroShare resource. (for example: https://www.hydroshare.org/resource/27045581bdea4808a393330f2417379c/)
2. Click Open With (top-right of the resource page) and select CIROH-2i2c JupyterHub.
3. Click “Login to continue” and authenticate with your CIROH credentials. (If this is your first time, you may be prompted to authorize access through GitHub.)
4. On the Server Options page:
   • Select the desired compute profile (Small, Medium, Large, or Huge).
   • Select the CIROH Community NextGen Hub image.
5. Click Start to launch the JupyterHub.
6. (If you did this with the example resource above, click on and launch one of the Jupyter Notebooks to step through the NextGen workflows.)

References​

• Nassar, A., Tarboton, D., Baig, F., Cunningham, J., Patel, A., Halgren, J., Lee, B., Salehabadi, H., Castronova, A., and Garousi-Nejad, I. "Accelerating community research applications of the NextGen water modeling framework through CUAHSI HydroShare linked with CIROH-2i2c JupyterHub." Environmental Modelling & Software. (In-review)
• Nassar, A., Tarboton, D., Baig, F., Patel, A., Cunningham, J., and Salehabadi, H. (2026). NextGen Water Modeling Workflow for Research-Scale Applications. https://www.hydroshare.org/resource/27045581bdea4808a393330f2417379c
• I-GUIDE VCO: Enabling Community Research with the Next Generation National Water Modeling Framework
• CIROH Hub: https://hub.ciroh.org/docs/products/ngiab-ecosystem/ngiab/distributions/nextgen-2i2c/

Having proven viability for daily timescale predictions on high-resolution river networks [4], MHPI researchers recently adapted δHBV 2.0 into a multi-timescale architecture designed to parameterize HBV and simulate streamflow at hourly intervals, at scale, across the NextGen HydroFabric. This new model, δHBV 2.0 MTS (Multi-TimeScale) [5], is a fusion of a daily and hourly δHBV 2.0 model designed to efficiently handle ML training with high geospatial and temporal complexity. (See MTS Architecture for more details about this construction.)

With δHBV 2.0 MTS maintaining similar forecasting skill compared to its daily-scale counterpart [5], Penn State and AWI were once again reunited in a joint effort to embed hourly scale differentiable modeling within AWI's operational ecosystem as a demonstration of model viability and to facilitate open access to its runtime.

Differentiable Models​

δHBV 2.0 and δHBV 2.0 MTS differentiable model constructions are briefly outlined here to contextualize the development efforts. For further detail, see each model's respective citation. At their core, differentiable models embed traditional process-based equations (here, the HBV rainfall-runoff model) inside a machine learning training loop. Because these models are designed to be differentiable (e.g., in PyTorch), gradients flow end-to-end from the loss function back through the physical equations and into the neural networks that supply their parameters. This lets the model learn optimal parameterizations directly from observed data while still obeying mass-balance and storage constraints encoded in HBV -- combining interpretability and physical consistency of process-based hydrology with the flexibility of deep learning.

δHBV 2.0​

The daily model uses Long Short-Term Memory (LSTM) and Multi-Layer Perceptron (MLP) neural networks (NNs) to learn time-dynamic and static parameters for HBV 2.0 [4] (a modified form of HBV). Weather forcings (precipitation, temperature, PET) and static catchment attributes are used as inputs to simulate hydrological states and fluxes:

where:

• $\theta$: Learned dynamic ($d$) and static ($s$) parameters.
• $x_m, A_m$: Forcings and attributes for unit basin $m$.
• $Q, S$: Model fluxes (e.g., streamflow) and states (e.g., snowpack).

In NextGen, hourly input forcings are aggregated at the end of every day to produce a single daily prediction, which is then distributed across the following 24 hours before a new prediction is made. Because HBV is a recurrent, bucket-based model, its internal states (e.g., groundwater stores) must be "warmed up" by running the model over a period just prior to the target simulation window so that storages can equilibrate. The same process initializes the internal states of the parameterization LSTM, which is itself recurrent.

MTS Architecture​

Extending δHBV 2.0 from daily to hourly resolution is not simply a matter of shortening the timestep. Hourly simulation demands sensitivity to rapid forcing variability, yet the model's internal storages still depend on slow-evolving, seasonal-scale memory. Training a single hourly model with a sufficiently long lookback to capture both regimes (i.e., 365 days) would be prohibitively expensive. The MTS (Multi-TimeScale) architecture [5] addresses this by coupling a daily and hourly δHBV 2.0 model in a two-stage design:

1. Daily warmup (long memory): The daily model processes ~351 days of aggregated daily inputs, building up slow-evolving states such as deep groundwater and snowpack.
2. Hourly warmup (short memory): Those states are handed off to the hourly model, which refines them over 7 days of hourly forcing to capture fast dynamics like sub-daily storm response.
3. Hourly simulation: With states fully initialized, the hourly model produces 7 days of predictions.
4. Rolling window: At the end of each 7-day simulation window, the cache shifts forward 7 days and the warmup process repeats.

A rolling-window input caching mechanism underpins this cycle, storing ~351 days of daily inputs alongside ~7 days of hourly inputs so that each warmup-simulation pass can be executed efficiently without reprocessing the full history.

Building A Stronger Bridge​

Supporting δHBV 2.0 MTS in AWI's NextGen, NGIAB, and NGIAB Data-Preprocess required significant contributions and dialogue between both institutions.

From Penn State, Leo Lonzarich incorporated MTS into the existing dhbv2 NextGen module, which couples with the differentiable modeling framework δMG to build hybrid models [3]. This update for MTS architecture addressed three key challenges for NextGen deployment: time-sequential runtimes not traditionally supported by ML models that operate on all timesteps simultaneously, the absence of daily forcing data, and no available derived potential evapotranspiration (PET) forcing.

To address the first two challenges, a new ring buffer mechanism was added to the BMI module. As described above, both the LSTM and HBV require warmup with historical data; the ring buffer caches batches of input forcings so that states can be warmed up in bulk (more efficiently than step-by-step) before the model advances sequentially with the initialized states. The ring buffer also handles the fact that NextGen (NWM 4.0, CIROH-UA) are intended to provide an hourly forcing while δHBV 2.0 MTS requires daily inputs as well. In such a case, the buffer aggregates hourly forcings to daily scale on the fly during runtime.

The final challenge was replacing Hargreaves PET, one of the standard inputs for δHBV 2.0 MTS, which previously needed to be computed in advance of runtime. Hargreaves' method is a daily measure requiring a full day of observations -- fundamentally incompatible with the MTS BMI's sequential operation. To resolve this, a Penman-Monteith calculation was added within the BMI to derive PET on-the-fly.

At AWI, Quinn Lee and Josh Cunningham, with help from Benjamin Lee, spearheaded the architectural integration and operational evaluation of the δHBV 2.0 module within the NGIAB ecosystem. AWI's efforts centered on data orchestration, model runtime validation and BMI compliance, and systems integration.

AWI developed automated workflows within NGIAB Data-Preprocess to generate the inputs required by both the daily and MTS models, aligning high-resolution HydroFabric attributes with hourly AORC forcing data and ensuring that realization and routing configurations met δHBV 2.0 and NextGen expectations. With these inputs, AWI tested the δHBV 2.0 BMI runtime to verify that outputs were reproducible across systems and that Data-Preprocess was producing correct model behavior.

Perhaps the most substantial effort involved reconciling δHBV 2.0's dependencies with the existing NGIAB architecture for Docker image construction. This ultimately required refactoring δHBV 2.0 and its supporting packages (δMG, hydrodl2) as well as parts of NGIAB's Dockerfile for more efficient builds.

In sum, this collaboration ensured that δHBV 2.0 MTS did not remain a standalone model, but a fully containerized, "plug-and-play" component of the NextGen ecosystem, ready for deployment across diverse computing environments.

Looking Forward​

With both daily- and hourly-scale δHBV 2.0 models implemented in NGIAB and NGIAB Data-Preprocess, the next step is adoption into the NextGen Research DataStream (NRDS). This work is already scheduled, and we anticipate recurring runtimes of both models with near-realtime simulations on the Community HydroFabric available to CIROH collaborators in the near future. Differentiable model simulations will then be accessible through the NRDS Visualizer, where researchers and community members can explore, assess, and iterate on outputs in real time.

For NOAA-OWP, ongoing work will continue to support the integration of δHBV 2.0 MTS into their NextGen test bench for preliminary evaluation as a candidate for the NWM 4.0.

Differentiable routing products are also in consideration for operational deployment. For instance, differentiable Muskingum-Cunge, introduced by Tadd Bindas et al. [1] at Penn State, is actively being developed and could appear operationally as a dynamic parameterization module for T-Route or as a standalone routing module. Public development is ongoing in DDR.

AWI and Penn State's strong collaborative track record is anticipated to continue as both teams support and potentially integrate new models into their NGIAB-NRDS ecosystem. Some of these efforts may also appear in DevCon later this spring. Ultimately, these successes stand as testaments to the value of inter-institutional collaboration for accelerating the integration of lab-bound research into operations for the broader CIROH community.

Code​

• δHBV 2.0 Operational Module Repository
• δMG Differentiable Modeling Framework (Backbone for δHBV 2.0; includes documentation and tutorials for building custom differentiable hydrologic models.)
• Differentiable Distributed Routing (DDR)
• MHPI Codebase

Publications​

1. Bindas, T., Tsai, W.-P., Liu, J., Rahmani, F., Feng, D., Bian, Y., et al. (2024). Improving river routing using a differentiable Muskingum-Cunge model and physics-informed machine learning. Water Resources Research, 60, e2023WR035337. https://doi.org/10.1029/2023WR035337
2. Patel, A., Halgren, J., Wills, Z., Frazier, N., Lee, B., Cunningham, J., et al. (2025). NextGen In A Box (NGIAB): Open-Source containerization of the NextGen framework to enable community-driven hydrology modeling. Environmental Modelling & Software, Volume 193. https://doi.org/10.1016/j.envsoft.2025.106666
3. Lonzarich, L., Shen, C., Song, Y., Rahmani, F., & Lawson, K. (2025). mhpi/generic_deltamodel: High-resolution differentiable model, 𝛿HBV2.0 (Version v1.4.0) [Computer software]. Zenodo. https://doi.org/10.5281/zenodo.14828008
4. Song, Y., Bindas, T., Shen, C., Ji, H., Knoben, W. J. M., Lonzarich, L., et al. (2025). High-resolution national-scale water modeling is enhanced by multiscale differentiable physics-informed machine learning. Water Resources Research, 61, e2024WR038928. https://doi.org/10.1029/2024WR038928
5. Yang, W., Ji, H., Lonzarich, L., Song, Y., Shen, C. (2025). Diffusion-Based Probabilistic Modeling for Hourly Streamflow Prediction and Assimilation. arXiv. https://arxiv.org/abs/2510.08488 [In Review]

To address this, we are excited to introduce CIROH Hub: a fully unified platform that brings all CIROH resources together in one place.

Visit hub.ciroh.org to explore the full breadth of CIROH’s research and resources, including dynamic collections of community datasets, applications, courses, and notebooks through HydroShare, a listing of CIROH publications through the community Zotero collection, extensive documentation on CIROH’s tools and software, guidance on accessing leading infrastructure for hydrological computing, and much more.

What's New in CIROH Hub​

CIROH Hub preserves everything you relied on from CIROH Portal and DocuHub, while introducing powerful new features to enhance visibility, usability, and collaboration across the consortium.

A Fresh Visual Experience​

CIROH Hub launches with a dynamic new visual theme with interactive elements. The interface is designed to be engaging for first‑time users while improving navigation for returning visitors.

Dedicated Jupyter Notebooks Portal​

Previously, Jupyter Notebooks were posted under the “Datasets” section of the CIROH Portal — a location that didn’t reflect their purpose and made them harder to find. CIROH Hub now offers a purpose‑built “Notebooks” section, giving notebooks a clear identity and improving their accessibility. Notebook authors can have their resources surfaced by adding the keyword ciroh_hub_notebook in HydroShare.

Public Directory of Working Groups​

CIROH Hub now also includes CIROH's Working Groups. We hope this resource will make it easy for researchers to identify working groups relevant to their work, enhancing collaboration and strengthening the consortium’s shared mission.

Contributing to CIROH Hub​

With all of the CIROH resources consolidated in one place, contributing your work has never been simpler. CIROH Hub enables you to share your resources with the community.

Share Research via HydroShare​

The “Research Portal” section of CIROH Hub hosts content from CIROH Portal and, like it, is open to community contributions. Whether you’d like to share a web application, dataset, educational course, presentation, or Jupyter Notebook, any of these resources can be easily surfaced onto CIROH Hub from HydroShare via the inclusion of appropriate keyword tags and optional metadata.

Update HydroShare Tags​

To coincide with the merger, new HydroShare tags have been established to populate resources in CIROH Hub. While we expect to support the existing, deprecated set of keywords indefinitely, we recommend transitioning your resources to use the new CIROH Hub keywords where possible.

A full list of current and deprecated HydroShare keywords is available on the Contribute page. Please ensure the appropriate tags are applied to your resources in HydroShare.

Add Publications via Zotero​

Submitting your publications via Zotero ensures they appear in CIROH Hub’s consolidated publication listings, which allows the community to view, iterate on, and cite your research. Request access to the CIROH group library now to get started.

Contributing Documentation​

If you’ve created a tool or software with useful documentation, the Documentation section of CIROH Hub is a great place to make that information easily accessible. This documentation can be automatically mirrored from your project’s GitHub repository README.md, allowing you to meet CIROH’s open-science best practices while keeping maintenance simple. To get started, submit a request or PR through the CIROH Hub repository.

Sunsetting CIROH Portal and DocuHub​

With CIROH Hub taking over their roles, we will be winding down CIROH Portal and DocuHub in the coming weeks. As of Monday, February 16th, 2026, neither site will be updated with the latest content and features, though dynamically loaded content will continue to populate both sites. During this time, we strongly recommend updating your links and bookmarks to point to the new site to ensure continuity of service for your users and readers.

Both CIROH Portal and DocuHub will remain online until Thursday, March 25th, 2026. When that date arrives, both portal.ciroh.org and docs.ciroh.org will be converted into redirect domains that point to the corresponding locations on CIROH Hub.

Accessibility Updates​

While CIROH Hub’s visual refresh may be appealing, we are aware that the site’s new animated elements may pose challenges for users with visual focus or computer performance issues. As such, we plan to introduce a reduced-motion mode by the end of March to address these limitations, ensuring that CIROH Hub’s core pursuit of accessibility is truly honored in every sense of the word. This timeframe means that both CIROH Portal and DocuHub will remain available in their current forms until a fully accessible version of CIROH Hub is available, ensuring that everyone can continue to take full advantage of CIROH resources.

Looking Ahead​

CIROH’s strength lies in its community and dedication to open science. By centralizing access to CIROH’s resources, we aim to bring the consortium into closer collaboration, maximizing our effectiveness as we translate cutting-edge advancements into operational successes in water management and public safety.

If you have any questions, feedback, or feature requests, reach out to the CIROH Hub team:

GitHub: Submit an issue on the ciroh_hub repository

Slack: Join the #ciroh-hub channel (Email ciroh-it-support@ua.edu for access)

Where Science Connects Us​

All good event attendees embrace the theme, and we happen to be great event attendees. This year's AGU theme, "Where Science Connects Us," aligned with CIROH's mission to democratize water modeling and foster community engagement in hydrologic research. Our presentations highlighted how we use containerization, cloud infrastructure, and collaborative frameworks in the world of operational and research hydrology. We also loved connecting with the broader CIROH community through the CIROH Research Exchange (lovingly dubbed T-REX by our talented events team), our biggest exhibit hall booth to date, a very packed poster presentation caravan led by our fearless Steve Burian, Martyn Clark, and Katie van Werkhoven, and, in classic AGU fashion, running into colleagues randomly in the extremely long convention center hallways.

Featured Presentations & Posters by CIROH Science Team & Friends​

NextGen In A Box (NGIAB) Ecosystem​

1. Advancing Hydrological Modeling: CIROH’s NextGen In A Box (NGIAB) and Enhanced Tools for Community-Driven Research

   Presenter: Bhavya Duvvuri

   Session: Poster Presentation H41Q-1427

2. NextGen Research DataStream: Community Contributions Towards Improved Hydrologic Predictions

   Presenter: Zach Wills

   Session: Oral Presentation H31G-03

3. Assessing the use of Physics and AI-Driven Precipitation Forecasts for Streamflow Prediction Using the NextGen In A Box (NGIAB) National Water Model

   Presenter: Jorge Bravo

   Session: Poster Presentation H41Q-1410

4. NextGen In A Box (NGIAB) and DataStream Visualizer: A Containerized Web Application for Interactive Geospatial and Temporal Visualization and Analysis of Hydrologic Model Outputs

   Presenter: Gio Romero

   Session: Oral Presentation H31G-08

5. Streamlining Community Modeling through Automated Data Processing for the Next Generation Water Resources Modeling Framework

   Presenter: Sonam Lama

   Session: Poster Presentation IN33B-0380

Cyberinfrastructure​

1. Cloud infrastructure for NextGen Water Resources Modeling: CIROH’s R2OHC Cloud Platform

   Presenter: Harsha Vemula

   Session: Poster Presentation H41Q-1425

2. High-Performance Computing Infrastructure for NextGen Water Resources Modeling: CIROH's R2OHC HPC Platform

   Presenter: Trupesh Patel

   Session: Oral Presentation H33J-07

Data Sharing & Community Engagement​

1. CIROH DocuHub and Portal: Knowledge Management System for Collaborative Water Resources Research

   Presenter: Manjila Singh

   Session: Poster Presentation H41Q-1426

2. Demonstrating new tools and frameworks to facilitate continental-scale hydrologic data access and evaluation

   Presenter: Sam Lamont

   Session: Oral Presentation IN42A-04

3. Enabling Community Research with the Next Generation National Water Modeling Framework on through Streamlined Workflows

   Presenter: Ayman Nassar

   Session: Oral Presentation H31G-07

4. Modernizing HydroShare: Using a Cloud-Native Architecture to Accelerate Water Science Innovation

   Presenter: Tony Castronova

   Session: Oral Presentation IN42A-02

Reflections from AGU25​

AGU is a wild world. It’s incredible how you can pack the population of a small town into one convention center and still have room left to spare. It’s even more incredible that at seemingly every turn, we always recognized at least a few citizens of said small town. We had such a wonderful time getting to see old friends and make new ones and come away from each conversation with a new idea for how CIROH Science Team could collaborate with the community. Our team was so encouraged by the number of presentations and posters we saw that used our community resources like our NGIAB container and our high-performance computing cyberinfrastructure offerings like Pantarhei and Wukong. (Perhaps you were one of our 13,101 NGIAB Docker deployments?) Not mentioned in the brief list above are all the presentations we got to see about model developments for the NextGen Framework. We love supporting our member and partner institutions’ research-to-operations work with our cyberinfrastructure resource offerings and community resource repositories like the Data Preprocessor, TEEHR, and the Tethys Visualizer. We would also love to showcase that work in our NextGen Research Datastream, running every hour for all of CONUS on AWS. If you’ve got a model that you’d like included in NGIAB and/or NRDS, we are more than happy to chat about it (or read through your GitHub issues submitted through our model integration issue form). We’re excited to get to work this year and support you and all of your R2O dreams!

Community Resources​

All AGU25 presentation materials, including posters and slide decks, are available in our GitHub repository: CIROH-UA/Conferences/AGU25

Join the Community and Get Involved​

Did AGU leave you wanting more from CIROH? We might not be able to see all of you in person every day (we miss y’all already), but whether you're a hydrologist, software developer, data scientist, or agency partner, there are many ways to get involved.

• Try NGIAB: Download from DockerHub or access NGIAB 101 for more details.

• Contribute Models to NGIAB: Submit BMI-compliant models integration request through our model integration issue form under the “New issue” button.

• Calibrate Models: Use our calibration tools to calibrate models for your favorite watersheds (we all have those, don’t pretend like you don’t know what we’re talking about!).

• Share Feedback: Open GitHub issues to report bugs and suggest improvements, or join our Community Resources Working Group discussions.

• Slack Channel: Join our public slack channels for community discussions

• Working Groups: Participate in one (or more, for you overachievers) of CIROH’s working groups – Artificial Intellgience, Community Resources, Decision Support, Early Career, Flood Inundation Mapping, Hydroinformatics, Hydrologic Modeling, Model and Forecast Evaluation.

• Student Summer Education Programs: Are you/do you know any undergraduates interested in AI and operational hydrology who are looking for research experiences over the summer? We have an REU for exactly that. Applications are due March 1. (Side effects: students may leave as Alabama football fans. Roll Tide!!)

• DevCon: Attend yearly Developer Conference in May. DevCon 2026 is in Salt Lake City, UT from May 27-29.

Acknowledgments​

CIROH's work is supported by NOAA Cooperative Institute Program under award NA22NWS4320003. We gratefully acknowledge the contributions from our numerous academic and agency partners who make this collaborative research possible.

The AGU25 presentations represent the collective effort of researchers, developers, and operational forecasters working together to advance hydrologic science. They are a culmination of months and years of dedication and love for science and research. Thank you to everyone who said hi, visited our posters, attended our presentations, and engaged in discussions about the future of water modeling. We’ll see you next year in San Francisco!

AI, DevOps and the Future of Cloud Infrastructure​

AWS re:Invent did not disappoint! I spent the first week of December at Amazon Web Services' flagship conference in Las Vegas. The event delivered cutting-edge technical insights, showcased the rapid evolution of cloud computing and AI, and provided countless opportunities to connect with industry leaders.

The energy across all five conference venues was more vibrant than I ever imagined it would be.

The Central Theme: AI and Agentic AI​

Unsurprisingly, the hottest topics throughout the conference were AI and Agentic AI. From keynote sessions to hands-on demos, the focus was on real-world applications, responsible AI development, and the next frontier of autonomous systems. Seeing how organizations are leveraging AI to create value sparked new ideas for advancing research infrastructure.

However, an important reality: security remains paramount. As Agentic AI systems grow more capable, robust security frameworks will be essential to ensure trust and stability.

A Developer and DevOps Paradise​

As a first-time attendee of AWS re:Invent, I was amazed by the work showcased during this conference! The conference showcased new ways to develop and bring ideas from concept to production-scale prototypes. As AWS CEO Matt Garman said in his keynote: "Why not?" That spirit of possibility and rapid innovation was there throughout the week. For those who missed it, I highly recommend watching the rapid-fire announcements from Matt’s keynote here: AWS re:Invent 2025 Keynote.

The keynotes from both the CEO and Dr. Swami Sivasubramanian were insightful. AWS is making strategic moves that are fundamentally changing how we deploy applications on the cloud—making workloads more cost-optimized and efficient. For CIROH, where we manage a hybrid infrastructure across multiple cloud platforms and on-premises, these insights are directly applicable to our ongoing efforts to democratize access to computational resources while managing costs effectively.

Community Connections​

One of my favorite experiences was attending the AWS Community Builder Hub. As a member of the AWS Community Builders program, I had the privilege of connecting with others working with AWS. The group was warm, welcoming, and incredibly insightful. Everyone I spoke with shared advice and perspectives that left me feeling both energized and supported.

I also got a chance to meet Jason Dunn in person—a great leader in sharing AWS content and community knowledge! These personal connections remind me why community-driven initiatives are so powerful.

Key Takeaways for Research Cyberinfrastructure​

The pace of technological advancement is staggering—tasks that once took months can now be completed in days thanks to AI, GenAI, and agent-based systems. But speed brings responsibility: security must scale with innovation.

The conference highlighted several critical areas:

• Cloud cost optimization & Storage innovations: Check out AWS Graviton5 Processors for HPC Workloads (could be a fit for those hydrological modeling workflows) and Amazon S3 Express One Zone for Research Data (great for frequently accessed datasets with new cost savings).
• Developer velocity: Lambda Durable Functions could simplify some of your long-running pipelines, and the new AI-powered workflow design for Step Functions makes building orchestrations much faster. You can use natural language to do this, so it starts to democratize Step Function development. You don't need to be a developer to use it.
• DevOps tooling: CDK Pipelines v2 with AI-Assisted Development has some cool features for infrastructure-as-code that could speed up your deployment cycles.
• Observability: Amazon CloudWatch AI-Powered Anomaly Detection, especially the cost anomaly detection feature, seems like a must-have for grant-funded budgets—it'll alert you when spending patterns deviate unexpectedly.

These developments reinforce CIROH’s hybrid approach—combining cloud scalability, collaborative computing, and on-premises HPC infrastructure-to stay ahead of the curve.

Learn More​

For deeper insights into AWS re:Invent 2025 announcements, check out these resources:

• AWS News Blogs: Top annoucements of AWS re:Invent 2025

• AWS Weekly Roundup: AWS re:Invent keynote recap, on-demand videos, and more

• A Journey into the cloud: From an on-prem admin to an AWS enthusiast

• re:Watch: AWS re:Invent Video Library

Last week, at the invitation and expert coordination of James Halgren, teams from RTI International (Sam Lamont and Matt Denno) and the University of Calgary (Darri Eythorsson, Cyril Thebault, and Martyn Clark) met at AWI for an intensive working session focused on weaving recent CIROH research into AWI’s fork of the NOAA Office of Water Prediction (OWP) Next Generation Water Resources Modeling Framework (nicknamed “NextGen”). James took the lead in developing the agenda, lining up the right scientific and technical expertise and ensuring that the week targeted the most critical software integration challenges. Throughout the visit, the RTI and UCalgary teams collaborated closely with AWI software engineers Quinn Lee, Josh Cunningham, hydrologic scientist Sifan A. Koriche, and James himself. The days were filled with whiteboards, deep technical conversations, and strategic planning around the future of NextGen water prediction. This recap captures the key themes and the momentum that carried through the week.

Getting started​

The U Calgary team arrived quite late Sunday night. The plane landed around 9:30 p.m., but with no dinner on board, the drive from Birmingham to Tuscaloosa became an unexpectedly comprehensive tour of places that were neither open nor in possession of a single vegetable. The group checked into the Indigo just before midnight, somewhat hungry but otherwise intact.

After Team Calgary’s epic late-night journey, we kicked off early on Monday morning with a whiteboard session to map out the “assets” we rely on for hydrologic modeling—datasets, models, tools—everything from the CAMELS dataset to TOPMODEL to NOAH-OWP-Modular, plus the assortment of geospatial processing and calibration methods that we all depend on. The aim was to start corralling this sprawling collection of community resources into something a bit more coherent, mostly because no single group has the expertise (or the workforce) needed to tackle all the challenges of building state-of-the-art prediction systems. The conversation naturally focused on software integration—the whole point of the trip—as we tried to figure out how to get these pieces to play nicely with each other.

A high point of the morning was meeting with our operational colleagues at the NOAA Office of Water Prediction—Dr. Fred Ogden, Dr. Fernando Salas, and Dr. Jason Regina. They shared updates on OWP’s vision for future versions of the National Water Model and outlined plans for defining model configurations and calibratng them across the national domain.

Refueled by a terrific lunch at the Curry Kitchen, we reconvened to focus on two key integration efforts: introducing the SYMFLUENCE integration architecture for large-domain modeling and enabling SUMMA’s process-based simulations to run seamlessly within the NextGen workflow.

Darri Eythorsson walked the group through SYMFLUENCE, an integration architecture designed to make hydrologic modeling workflows more structured and efficient, describing how SYMFLUENCE supports everything from defining spatial domains and preparing data to model setup, calibration, optimization, and evaluation. He emphasized that SYMFLUENCE can also drive simulations from NextGen, giving NextGen access to capabilities such as large-sample calibration and geospatial preprocessing. Darri noted that the system runs on laptops, HPC clusters, and cloud-compute platforms, making it adaptable to modeling at different scales.

We closed out Monday with a deep dive into bringing SUMMA into NOAA/OWP’s NextGen framework. The meeting was scheduled late in the day so that Ashley van Beusekom, one of SUMMA’s primary developers, could join from her home in Okinawa, Japan. A couple of years ago, Ashley added BMI support to SUMMA, making it possible to run the model within the NextGen architecture. Ahead of the meeting, she and Darri Eythorsson worked together to set up SUMMA as a spatially distributed model in NextGen, and they successfully reproduced the Provo and Oyster River test cases to verify that everything was running as intended. Quinn and Josh from AWI were excited by the results and immediately started weaving SUMMA capabilities into the broader NextGen framework.

The U Calgary crew wrapped up the day on the Indigo Hotel rooftop, sharing giant pretzels, vegetables, and the warm Alabama night air. They toasted the momentum of the visit and the possibilities ahead. As the night settled in, Team RTI’s Matt Denno and Sam Lamont arrived mid-evening Monday, ready to jump into the work the next morning.

Technical deep dives​

The RTI and U Calgary teams met up for breakfast on Tuesday morning at Heritage House where they exchanged high-fives and consumed the caffeine and carbohydrates needed to tackle another ambitious day of software integration.

Once everyone settled in at AWI, the group dug into the two main challenges of the week: integrating models into the NextGen framework (the U Calgary effort) and evaluating those simulations with shared protocols (the RTI focus). The conversation broadened as Arpita Patel (UA), Jordan Laser (Lynker), Zach Wills (Lynker), Nels Frazier (Lynker), Sifan A. Koriche (AWI), Giovanni Romero (Aquaveo), Jonathan Frame (UA), Trupesh Patel (UA), Harsha Vemula (UA) and Manjila Singh (UA) joined to discuss their work on making NextGen accessible and useful to the wider community. Together, the group reviewed progress on SYMFLUENCE, the NextGen/SUMMA integration, and the TEEHR data model. As always, James Halgren kept everything running seamlessly, turning lunch into a productive working session so the team could update the University of Alabama’s computer science leads on recent advances and ongoing challenges.

Before long, the team divided into smaller groups so they could move quickly on the most pressing software-integration tasks on the schedule. Throughout the day, the big question hovered: How do we create a seamless pipeline from model execution data ingest→ evaluation → model improvement?

On the RTI side, much of the effort focused on preparing outputs for seamless use within the TEEHR Evaluation System. This involved translating the output from NextGen simulations into the TEEHR data schema and determining how best to extract the metadata needed to track which model configuration produced each forecast. The team also worked on defining crosswalks to link simulation points to USGS gage IDs, ensuring that model results could be cleanly tied back to the appropriate metadata and hydrofabric, and they began documenting key operational details such as run intervals and latency characteristics of near-real-time runs. By the time the dust settled, the RTI and AWI teams had stitched the pieces together and emerged with the gratifying sense that they had made good initial progress on the problem they came to solve.

On the U Calgary side, the team split off into two groups. In one corner Darri Eythorsson and Cyril Thebault met with Josh Cunningham to make progress on integrating SUMMA in to NextGen. There were lots of issues to consider, including appropriately linking with different software libraries and ensuring that SUMMA could run effectively in the containerized NextGen instantiation “NextGen In a Box” (NGIAB). The software integration progress was rapid, thanks in large part to the Herculean efforts of Ashley van Beusekom prior to the visit.

In another corner, Martyn Clark sat down with James Halgren, Quinn Lee and Sifan A. Koriche to put together a SWOT (Strengths, Weaknesses, Opportunities, Threats) analysis for the AWI-led efforts to transition CIROH research into NOAA/OWP operations. The timing mattered: James was preparing to drive to Utah the following week for a faculty position at Brigham Young University, and leadership of his group was in the process of transitioning to Quinn Lee and Josh Cunningham. The SWOT analysis brought a sense of optimism and excitement as the team saw the promise in recent hydrologic prediction advances and brainstormed several clear steps to make the research-to-operations path smoother and more effective.

Later in the day, Arpita Patel and Matt Denno met with NOAA/OWP Chief Scientist Fred Ogden to walk through the latest NextGen Framework design diagram and discuss how their development of NextGen In A Box (NGIAB) aligns with NOAA/OWP’s component-based architecture. Their conversation highlighted that NGIAB already follows many of these principles, including modular components, stand-alone tools, and fully BMI-compliant model execution. They also identified clear opportunities to strengthen the alignment, particularly around refining the runtime environment, enhancing component decoupling, and mapping NGIAB’s architecture onto OWP’s emerging design. The meeting set the stage for ongoing coordination.

Dinner at the River Café was a highlight of the day. With a table on the terrace and the Black Warrior River drifting by below, the team enjoyed an evening of easy conversation and good-natured reflection on the progress over the past two days.

A springboard for the future​

Wednesday saw the RTI and U Calgary teams peel off in different directions. RTI paired with AWI software engineers to advance the TEEHR model evaluation manager, tightening the evaluation protocols, metadata requirements, and workflow linkages needed for basin-scale and national-scale verification. U Calgary spent the day with CIROH leadership —Steve Burian, Sagy Cohen, and Erin White—reviewing their major research advances and the steps required to improve the research-to-operations pipeline. A clear highlight of the day came when SUMMA was successfully integrated into NextGen: Martyn Clark marked the moment with a celebratory dance—hands clasped, fingers interlocked, and his upper body swaying with unmistakable satisfaction.

That evening, everyone gathered at Steve Burian’s home for a farewell celebration in honor of James Halgren. Steve and his wife Cindy laid out an extraordinary taco spread, enough to nourish the lively crowd that had gathered to see James off in style. James brought with him a tall stack of mounted photographs he had taken around the U.S., and he spent the night offering warm goodbyes and writing personal messages on the back of each one. It was a lovely gesture, and a reminder of how much his enthusiasm and kindness had meant to the community.

The joint visit wrapped up on Thursday morning, when the full team gathered to chart a path forward. Together they mapped out the key components of the NextGen Research Data Stream (NRDS—affectionately pronounced “nerds”) and clarified how NRDS fits within the broader NOAA/OWP NextGen framework, including the orchestration manager, calibration manager, and evaluation manager that had occupied so much attention throughout the week. The discussion concluded with a clear set of immediate priorities, identifying the specific steps needed over the next six months to realize near-term successes in the transition of research to operations.

CIROH Cyberinfrastructure: Accelerating Hydrologic Research​

This collaborative work exemplifies how CIROH's cyberinfrastructure investments are accelerating the pace of hydrologic research. The platforms developed by CIROH's DevOps team - the Research to Operations Hydrologic Community (R2OHC) platform and tools developed by CIROH's Science team, Lynker team, RTI team and Aquaveo team —including NextGen In A Box (NGIAB), NRDS, and containerized modeling environments—provide researchers with immediate access to computational resources and standardized workflows. By eliminating the traditional barriers of infrastructure setup and configuration, these platforms allow research teams like University of Calgary, RTI International, Lynker, Aquaveo and AWI to focus their energy on scientific innovation rather than technical implementation. This meeting at AWI represents the kind of rapid, collaborative progress that becomes possible when robust cyberinfrastructure removes technical friction from the research process.

Almost from the start, 2025 has been a banner year in hydrologic modeling, with advancements in capabilities on both sides of the aisle of CIROH's research-to-operations (R2O) pipeline.

• From the research skunkworks, Penn State's MHPI group, led by Dr. Chaopeng Shen introduced a new generation of distributed, differentiable hydrologic models spearheaded by δHBV 2.0. Capable of high-resolution, continental-scale streamflow forecasting across the CONUS Hydrofabric, δHBV 2.0 fuses process-based modeling and machine learning to enable efficient parameter calibration and interpretable predictions at scale -- with demonstrated viability as a National Water Model 3.0 successor.

• Meanwhile, from the operations core, CIROH's Science and Cyberinfrastructure team at the Alabama Water institute (AWI), headed by Arpita Patel and James Halgren, debuted an operational pipeline composed chiefly of flagships NextGen In a Box (NGIAB) and the Next Generation Research Datastream (NRDS), forming an open-source software stack powering an accessible operational implementation of NOAA's Next Generation National Water Model. NRDS, in particular, executes a regularly scheduled set of continental-scale, NextGen-based hydrologic forecasts on CIROH's cloud-based cyberinfrastructure, and is designed to be a canvas for showcasing community modeling advances as potential R2O opportunities for NextGen.

As destined in CIROH, these two efforts converged in early October during a week-long collaboration hosted by AWI, bringing Penn State researcher Leo Lonzarich south to Tuscaloosa, Alabama, to join in facilitating the formal introduction of differentiable models like δHBV 2.0 into the NRDS.

Keyboard Accomplishments 👾​

During the visit, the CIROH Science and Cyberinfrastructure team worked with Leo to finalize the formal integration of the δHBV 2.0 model into the NextGen ecosystem. Between extensive code reviews, model refactoring, and validation runs, we verified δHBV 2.0's operations-ready performance through its Basic Model Interface (BMI) within the NextGen runtime, ensuring full compliance with BMI standards and alignment with previously published benchmarks. We enhanced the model to support both batch and sequential timestep simulations — the latter mirroring the runtime behavior of NextGen — while maintaining high computational efficiency and reproducibility across environments.

Once δHBV 2.0 was verified and tuned, we expanded the scope to broader ecosystem integration. The model was finalized on GitHub with modular configuration capabilities, daily and hourly (model coming soon) simulation modes, and confirmed compatibility with the T-route routing component. Significant performance optimizations were implemented to improve scalability across large basin networks. We added δHBV 2.0 support to the NGIAB distribution, enabling cloud-based state loading from AWS S3 (PR coming soon), and extended its reach into the NGIAB Data Preprocess to allow on-demand generation of forcings, static catchment attributes, and model realizations for NextGen. Finally, the static catchment attributes were incorporated into the CIROH Community Hydrofabric, ensuring the model is represented and interoperable within the hydrologic framework used across the consortium.

The beauty of these efforts lies in the fact that δHBV 2.0 is built upon a model-agnostic, differentiable modeling framework — δMG. Therefore, the formalization accomplished during this visit scales to all new research products and advancements made through the framework, streamlining future NGIAB and NRDS integrations and reducing development overhead across projects.

Beyond the Keyboard​

Out of the office, this week also offered opportunities for cross-team engagement and a deeper look into CIROH's computational backbone. In one such case, Leo and the team toured the University of Alabama's data center, meeting with Josh Lotfi and other members of the Office of Information Technology (OIT) who help manage CIROH's on-premises high-performance computing resources. Naturally, this included 1-on-1 time with the flashing lights and silicon of Wukong (which supports a bulk of Penn State's R&D) and Pantarhei HPCs. Across the board, these interactions gave valuable insight into work going on behind the scenes at AWI and about how CIROH's cyberinfrastructure supports researchers and partners across academia, government, and industry.

It is easy to see these collaborations as purely technical exchanges. However, the deep friendships that were built during this visit are at least as important, they inevitably inspire open flows of ideas, mentorship, and future synergies in service of CIROH's mission and the broader community.

Looking Forward​

With δHBV 2.0's integration near completion, differentiable models will soon be running as part of CIROH's NRDS nightly forecasts — accessible through the NRDS Visualizer where other researchers and community members will be able to explore, assess, and iterate on simulations from this latest crop of hydrologic models in real time.

We anticipate continuing this AWI-Penn State collaboration, with future projects e.g., being additions to NGIAB and NRDS including more differentiable models, a formal hourly δHBV 2.0 model, and differentiable routing. Some of this work may also arrive in a DevCon to showcase the efforts of both of groups.

Overall, this visit exemplifies how CIROH's approach to developing a community of practice catalyzes scientific advancements, fosters lasting inter-institutional relationships, and makes normally lab-bound research products salient to the broader R2O community. In these collaborations, CIROH's core mission, strengthening the link between scientific discovery and operational forecasting, remains clear and in view.

Many thanks are owed to Arpita, James, and the whole of CIROH's Science and Cyberinfrastructure team -- git merge on research products has never been easier.

Resources for the Curious​

• δHBV 2.0 Operational Model Repository

• δMG Differentiable Modeling Framework (Backbone for δHBV 2.0; includes documentation and tutorials for building custom differentiable hydrologic models.)

• MHPI Codebase

Publication​

1. Song, Y., Bindas, T., Shen, C., Ji, H., Knoben, W. J. M., Lonzarich, L., et al. (2025). High-resolution national-scale water modeling is enhanced by multiscale differentiable physics-informed machine learning. Water Resources Research, 61, e2024WR038928. https://doi.org/10.1029/2024WR038928

Meeting Highlights: Transforming Research into Real-World Impact​

Over four days, our community showed how scientific innovation translates into tools and systems that support communities, enhance resilience, and improve decision-making across the nation. From AI advances in water prediction to flood inundation mapping, the presentations and discussions demonstrated the collaborative, open-science approach that defines CIROH.

Access Meeting Resources​

We are excited to share a comprehensive collection of assets from the 2025 CIROH Science Meeting, including session recordings, presentation slides, and group photos! Whether you attended in person, joined virtually, or could not participate, these materials are available to you.

🔗 Access the Shared Drive​

Please note that materials are being uploaded continuously as we receive images and additional files from presenters and participants. We encourage you to check back periodically for updates. If you are a presenter and prefer not to have your slides or poster included in the shared drive, please contact Charity McCalpin (cnmccalpin@ua.edu) for removal.

🌐 Visit the Event Microsite​

Explore the agenda, highlights, and media on the CIROH Science Meeting site.

👉 Ciroh Science Meeting

Your Feedback Matters​

📝 Complete Our Post-Event Survey​

If you attended the meeting, either in person or virtually, we kindly ask that you complete our post-event survey. Your feedback is invaluable in helping us improve future meetings and ensure they continue to serve our community's needs effectively.

👉 Complete the Post-Event Survey

Looking Forward: Building on Our Collaborative Foundation​

The 2025 Science Meeting reinforced CIROH's commitment to building a collaborative, open-science community that transforms research into real-world impact. The connections made and innovations shared during these four days will continue to drive our mission forward throughout the year.

Thank You​

Thank you for your ongoing support and participation in making the 2025 CIROH Science Meeting a tremendous success. Your commitment to advancing water science and translating research into operational tools that benefit communities nationwide is what makes CIROH's mission possible.

Together, we continue building a future where scientific innovation directly serves society's water-related challenges and opportunities.

The CIROH Science Meeting 2025 was held September 15-18 in Tuscaloosa, Alabama. For more information about future meetings and CIROH initiatives, visit our website (https://ciroh.ua.edu/) and stay connected with our community.

With help from Arpita Patel and the CIROH Cyberinfrastructure and DevOps Team, who assisted our team with access to Amazon Web Services and the Google Cloud Platform, we developed and tested new cyberinfrastructure that advances camera-based hydrologic monitoring.

Traditional dataloggers used in hydrologic monitoring focus on interfacing with conventional sensors (e.g., pressure transducers, float gages, etc.) and lack some capabilities required for camera-based monitoring. Low-cost field computers like the Raspberry Pi provide a capable alternative, but lack out-of-the-box software required to support high-resolution image and video capture, management of the large volume of data that accumulates, data processing, and cloud uploading processes. Because of this, we had to build the functionality required to combine low-cost field computers with cloud computing services to produce an operational, real-time, cloud-integrated, camera-based streamflow monitoring system.

We developed two new Python-based software programs compatible with inexpensive field computers such as Raspberry Pi: one that captures and processes the data from cameras, and one that automatically extracts hydrologically relevant data from image data in the cloud.

1. Hydrocam Collect: automatically captures images and videos at user-specified intervals, manages local storage on field computers, and uploads captured image and video data to the cloud. Real-time data monitoring and alerts minimize potential for data loss.
2. Hydrocam Compute: This serverless cloud computing workflow extracts relevant hydrologic data from camera images and videos, such as depth/stage and velocity. Automated and scalable, the workflow can handle a large number of independent monitoring sites, each executing their own serverless functions using isolated cloud resources.

In an operational context, automation is key to a scalable, robust system capable of integrating camera-based monitoring with larger hydrologic monitoring networks such as those managed by the United States Geological Survey (USGS). We tested the automation of the system we developed using commercial cloud services via Amazon Web Services and Google Cloud Platform with access provided through CIROH. By leveraging CIROH-provided cloud services, our team was able to prototype, test, and scale the software to bridge the gap between proof-of-concept research and practical field use.

This linked software system we developed responds to data events (i.e., capture and upload of an image or video to the cloud) instantly and minimizes human intervention and error correction. The serverless and event driven design of Hydrocam Compute reduces infrastructure management overhead and cost because cloud resources are only used during execution of data processing tasks with no need to manage or maintain physical compute infrastructure. The models used for processing images and video can be swapped out to meet the needs of particular monitoring locations.

The Hydrocam workflows enhance the feasibility, reliability, and efficiency of camera-based streamflow monitoring. As hydrological conditions become more variable, access to high-resolution, real-time data becomes increasingly important for flood forecasting, water resource management, and long-term climate adaptation strategies. This work supports these broader goals in making camera-based monitoring more practical and scalable.

Resources and Publications​

Open Source Software​

The HydrocamCollect and HydrocamCompute software programs are available in GitHub repositories within the hydrocam GitHub Organization.

Publications​

Access the pre-prints of papers related to the Hydrocam software and camera-based monitoring:

1. Neupane, S., Horsburgh, J. S., Bin Issa, R., Young, S. (2025). HydrocamCollect: A robust data acquisition and cloud data transfer workflow for camera-based hydrological monitoring. Environmental Modelling & Software (submitted). Available at SSRN: https://ssrn.com/abstract=5451102 or https://doi.org/10.2139/ssrn.5451102
2. Neupane, S., Horsburgh, J. S., Bin Issa, R., Young, S. (2025). HydrocamCompute: Serverless cloud computing workflow for camera-based hydrological monitoring. Environmental Modelling & Software (submitted). Available at SSRN: https://ssrn.com/abstract=5451100 or https://doi.org/10.2139/ssrn.5451100
3. Bin Issa, R., Neupane, S., Khan, S., Horsburgh, J. S., Young, S. (2025). Towards Real-Time Water Level and Discharge Measurements using Imagery, Machine Learning, and Edge Computing. Journal of Hydrology (submitted). Available at SSRN: https://ssrn.com/abstract=5531964 or https://doi.org/10.2139/ssrn.5531964

Contact​

• Dr. Sierra Young, Utah State University, sierra.young@usu.edu
• Dr. Jeff Horsburgh, Utah State University, jeff.horsburgh@usu.edu

The National Weather Service's Middle Atlantic River Forecast Center (MARFC) sees large variations in the performance of the National Water Model 3.0. Through its support for regionalized parameters and models, NOAA-OWP’s Next Generation Water Resources Modeling Framework (NextGen framework) offers a potential solution to address these inconsistencies. As such, this study took advantage of NextGen in a Box (NGIAB) to evaluate the NextGen framework’s performance in the MARFC region.

This study evaluated three operational hydrologic modeling frameworks targetted at the National Water Model (NWM): the Community Hydrologic Prediction System (CHPS), the NextGen framework, and version 3.0 of the National Water Model itself.

• CHPS is the current operational framework used by NOAA's River Forecast Centers. It incorporates the SNOW-17 model for snowmelt and the Sacramento Soil Moisture Accounting (SAC-SMA) model for runoff generation.
• For the early phases of this study, the NextGen framework was used with the default model configuration provided by the NGIAB ecosystem, which combines the Noah-OWP-Modular land surface model and the Conceptual Functional Equivalent (CFE) rainfall runoff model [2].
  • After initial runs with the baseline configuration, Noah-OWP-Modular was replaced with SNOW-17 output and simplified Potential Evapotranspiration (PET) values from the MARFC database.
  • The models were calibrated using two objective functions: Kling-Gupta Efficiency (KGE) [6][7] and Nash-Sutcliffe Efficiency (NSE) [4][5].
• The National Water Model 3.0 uses the Noah-MP land surface model coupled with the Weather Research and Forecasting Hydrologic model (WRF-Hydro) [2][3] to simulate hydrological processes across CONUS.

The case studies focused on the Westfield and Elkland basins in North-Central Pennsylvania. These basins provide good locations for comparison due to the presence of USGS stream gages and their "flashy" behavior, characterized by rapid and unpredictable rises and falls in streamflow. Additionally, both Westfield and Elkland were sites of catastrophic flooding during Tropical Storm Debby in 2024, which allowed for the models to be evaluated on a recent extreme flood event. Results from Westfield, PA are shown in Figure 1.

The simulations executed for the case study are listed below:

• Uncalibrated CHPS
• Calibrated CHPS
• Uncalibrated NextGen (AORC forcings)
• Uncalibrated NextGen (MARFC forcings)
• Calibrated NextGen (AORC forcings; calibrated on KGE objective function)
• Calibrated NextGen (AORC forcings; calibrated on NSE objective function)
• Calibrated NextGen (MARFC forcings; calibrated on KGE objective function)
• National Water Model 3.0

These simulations were executed on the JetStream2 platform with the support of the CIROH Research Cyberinfrastructure Team.

The researchers observed that the calibrated runs for both NextGen and CHPS showed significant improvements over uncalibrated runs in both Westfield and Elkland basins. Among the calibrated runs, CHPS marginally outperformed NextGen. Additionally, both CHPS and NextGen outperformed NWM 3.0 in Westfield basin.

When running the NextGen-based models, substitution of SNOW-17 output and simpler PET for the NOAH-OWP reduced calibration time by 50% and produced similar outputs.

However, the models that performed best during calibration did not necessarily perform best during the Hurricane Debby extreme event. This suggests that including a greater number of extreme events in the calibration period is necessary to improve model robustness.

NGIAB served as a critical enabler in this study, allowing for rapid iteration and in-depth exploration of the NextGen framework’s configuration space. This evaluation of objective functions, forcing datasets, and baseline models would have been significantly more challenging with traditional workflows.

The presentation and output data for this study have been made available on Hydroshare.

References​

1. National Oceanic and Atmospheric Administration. Community Hydrologic Prediction System (CHPS) Documentation [Internet]. National Oceanic and Atmospheric Administration; [updated 2024 Nov 26; cited 2025 Jul 28]. Available from: https://vlab.noaa.gov/web/chps
2. Araki R, Ogden FL, McMillan HK. Testing Soil Moisture Performance Measures in the Conceptual‐Functional Equivalent to the WRF-Hydro National Water Model. JAWRA Journal of the American Water Resources Association. 2025 Feb;61(1):e70002.
3. National Oceanic and Atmospheric Administration. The National Water Model [Internet]. National Water Prediction Service; [cited 2025 Jul 29]. Available from: https://water.noaa.gov/about/nwm
4. Tijerina‐Kreuzer D, Condon L, FitzGerald K, Dugger A, O’Neill MM, Sampson K, Gochis D, Maxwell R. Continental hydrologic intercomparison project, phase 1: A large‐scale hydrologic model comparison over the continental United States. Water Resources Research. 2021 Jul;57(7):e2020WR028931.
5. Moriasi DN, Arnold JG, Van Liew MW, Bingner RL, Harmel RD, Veith TL. Model evaluation guidelines for systematic quantification of accuracy in watershed simulations. Transactions of the ASABE. 2007;50(3):885-900.
6. Gupta HV, Kling H, Yilmaz KK, Martinez GF. Decomposition of the mean squared error and NSE performance criteria: Implications for improving hydrological modelling. Journal of Hydrology. 2009 Oct 20;377(1-2):80-91.
7. Vrugt JA, de Oliveira DY. Confidence intervals of the Kling-Gupta efficiency. Journal of Hydrology. 2022 Sep 1;612:127968.

Recent collaboration between researchers in the Cooperative Institute for Research to Operations in Hydrology (CIROH) from University of Alabama (UA) and Utah State University (USU) highlighted the value of cross-institutional partnerships in improving community hydrologic modeling. Focused on the Logan River watershed, this joint effort demonstrated how sharing tools, knowledge, and infrastructure can accelerate both model development and scientific discovery.

Through this engagement, USU researchers gained deeper understanding of the NextGen framework and T-Route modeling library, empowering them to improve physical process representations for the Logan River watershed for heightened simulation fidelity. The collaboration also provided valuable exposure to the developmental side of complex modeling tools, offering insights into framework design, automation workflows, and best practices for model setup and calibration. Both teams benefited from exposure to alternative research tools and methods, which helped enhance and refine the community development pipeline.

One key outcome was the correction of a spatial representation in the Logan River hydrofabric (Figure 1) using the USU’s specific local knowledge. The reach for this region (highlighted in the zoomed subfigure) was updated to reflect its real world path, where it flows into an underwater tunnel that reconnects to Logan River further downstream. This correction will be integrated into the community hydrofabric, benefiting the wider community by improving model realism and reliability. Furthermore, in combination with freshly calibrated model parameters, these updates have improved the model's KGE metric by 0.2 units — a significant gain in model accuracy.

NGIAB played a central role in this success, streamlining the process of setting up, running, and analyzing multiple model configurations. By removing the typical installation and troubleshooting barriers, researchers could focus their time on advancing physical representations rather than infrastructure issues.

This collaboration sets a precedent for how community-driven, open science efforts can improve hydrologic model performance and usability. By combining localized knowledge with shared national tools, we move closer to creating reliable, reusable, and scalable flood forecasting systems, a crucial step for Research-to-Operations (R2O) success.

We look forward to repeating this model of collaboration with other regional experts and academic institutions to continually refine the National Water Model ecosystem and its supporting frameworks.

We're thrilled to announce that NextGen In A Box (NGIAB) has surpassed 10,000 Docker pulls — a significant milestone reflecting the growing adoption of water modeling tools that are accessible to all. This achievement creates opportunities for researchers, practitioners, and students worldwide to leverage advanced water prediction frameworks without infrastructure barriers, accelerating global water science innovation.

Update, 8/29: NGIAB Journal Paper now available in Environmental Modelling and Software
→ Read the full paper

From Research Innovation to Community Tool​

When we first containerized the NextGen Water Resources Modeling Framework into NGIAB, our goal was simple yet ambitious: remove the technical barriers that prevented many researchers from accessing NOAA's next-generation water modeling capabilities.

Today, with over 10,000 downloads, it's clear the community was ready for this transformation.

The University of Alabama recently highlighted NGIAB's impact in their news feature, "UA Software Makes Water Modeling More Accessible", recognizing how this tool is changing the landscape of hydrologic research and education. As the article notes, NGIAB turns what was once a complex, infrastructure-heavy process into something that researchers can run on their laptops in minutes.

What 10,000 Pulls Really Means​

Behind this number are stories of:

🎓 Graduate students exploring advanced modeling techniques with no setup headaches

🏫 Educators bringing cutting-edge tools into the classroom

🏢 Researchers at smaller institutions gaining access to national-scale modeling

🌍 International collaborators contributing to water modeling advancement without infrastructure constraints

🚨 Emergency managers rapidly deploying models for flood prediction

Community Growth and Impact​

The rapid adoption of NGIAB reflects a broader movement:

1. Democratization of Advanced Modeling
No longer do researchers need access to specialized HPC resources or deep DevOps knowledge to run sophisticated water models. NGIAB levels the playing field.

2. Reproducible Science
Every one of those 10,000 pulls represents the exact same computational environment, ensuring that research results can be replicated anywhere in the world.

3. Accelerated Innovation
By removing setup friction, researchers can focus on science rather than software configuration, leading to faster iterations and discoveries.

Looking Forward: The Next 10,000​

As we celebrate this milestone, we're already working on what's next:

📦 Expanded model library: Adding more BMI-compliant models to the NGIAB ecosystem

☁️ Cloud integration: Seamless deployment on AWS, Google Cloud, and Azure

📊 Enhanced visualization: Built-in tools for analyzing and presenting model outputs

🤝 Community contributions: Making it easier for users to share their configurations and improvements

Join the NGIAB Community​

Whether you're pull number 10,001 or have been with us since the beginning, you're part of a growing community that's transforming water resources modeling.

Here's how to get involved:

👉 Try NGIAB: Visit the NGIAB 101 tutorial to get started

✍️ Share your use case: Tell us how you're using NGIAB in your research or operations

🛠 Contribute: Submit bug reports, feature requests, or code contributions

📢 Spread the word: Help others discover how NGIAB can accelerate their water modeling work

Thank You to Our Community​

This milestone belongs to everyone who downloaded, tested, provided feedback, contributed code, or helped spread the word about NGIAB. Your engagement drives our continuous improvement and motivates us to make water modeling even more accessible.

As we reflect on reaching 10,000 Docker pulls, we're reminded that each download represents a researcher, student, or practitioner working to better understand and predict our water resources. Together, we're building a more resilient future through accessible, advanced water modeling.

Here's to the next 10,000 pulls and the continued growth of our community!

NGIAB is developed and maintained by the Cooperative Institute for Research to Operations in Hydrology (CIROH) at the University of Alabama. Learn more about our work in making water modeling accessible at ciroh.org.

CIROH team at NHWC 2025 in Tucson, Arizona, standing by the event’s official banner.

CIROH had a strong showing at the 15th Biennial National Hydrologic Warning Conference (NHWC 2025), with our researchers presenting innovative solutions and engaging with the broader hydrologic warning community. The conference brought together field personnel, innovators, engineers, hydrologists, forecasters, water resource managers, and emergency management officials from across the country to advance flood warning systems and address emerging challenges in evolving climate and drought management.

Advancing Model Reliability in Hydrologic Forecasting​

During the NextGen presentation track, moderated by Dr. Steve Burian, CIROH researchers focused on improving evaluation techniques for hydrologic models—particularly in the context of the National Water Model (NWM).

NWS NextGen Program​

Edwin Welles from Deltares USA presenting the NextGen framework.

Edwin Welles from Deltares USA kicked things off with insights into a groundbreaking initiative. The Office of Water Prediction, of the National Weather Service has undertaken an ambitious project to implement and apply a new framework for hydrological modeling: the The Next Genearation Water Resources Modeling Framework, or the NextGen framework, is designed to emphasize modularity, interoperability, and reproducibility, creating an environment where disparate modeling components can be freely combined to enhance modeling capabilities. These advantages offer critical support to the National Water Model, alongside other key initiatives such as national Flood Inundation Mapping research. The system is being implemented by Raytheon with subcontract support from Deltares USA. His presentation provided a summary of the project and envisioned outcomes.

Leading the NextGen Revolution​

Arpita Patel presenting “Navigating the NextGen Ecosystem and NextGen In A Box (NGIAB).”

Arpita Patel followed with “Navigating the NextGen Ecosystem and NextGen In A Box (NGIAB),” a presentation focused on making the NextGen framework more accessible to researchers, forecasters, and water managers. This was achieved through the NextGen In A Box platform, which harnesses containerization and supportive tooling to make the NextGen framework accessible and easy to use. This work directly supports CIROH’s mission to democratize access to advanced hydrologic modeling capabilities and foster broader adoption of next-generation water prediction tools.

🥉 Patel also demonstrated exceptional engagement throughout the conference. Her outstanding participation earned her an impressive #3 ranking on the NHWC 2025 conference leaderboard with 70,200 points. This demonstrated not only her technical expertise, but also her active involvement in conference activities and community engagement.

Rethinking Evaluation​

Dr. Md Shahabul Alam presenting different model evaluation methods.

Dr. Md Shahabul Alam also delivered a compelling talk titled “Rethinking Model Reliability: A Dual Evaluation of NWM Using Time Series and Extreme Events” during the NextGen program track. His work focuses on advancing operational hydrology by developing more rigorous and resilient methods to evaluate the National Water Model across varying temporal scales and extreme event conditions—a crucial step in improving forecasting and decision-making under a changing climate. This work represents CIROH’s commitment to enhancing the reliability and accuracy of national-scale hydrologic prediction systems, directly supporting the operational forecasting community’s need for trustworthy model outputs during both routine conditions and extreme weather events.

🥇 Dr. Alam earned also earned the #1 ranking on the NHWC 2025 conference leaderboard, maintaining a commanding lead of over 100,000 points throughout the week and finishing with a remarkable 194,700 points. His name quickly became a focal point of the conference. Each morning, attendees checked the Whova app with anticipation—only to see Shahab had once again pulled further ahead. His energetic participation and engaging presence made him a standout at the conference, combining technical depth with social and technological savvy. Dr. Alam not only demonstrated exceptional leadership and charisma at the NHWC 2025 conference, but also brought well-deserved visibility to CIROH and its ongoing mission.

Strengthening Community Connections​

The inclusion of these presentations within the NextGen presentation track highlights CIROH’s central role in advancing the NextGen Water Resources Modeling Framework. This alignment demonstrates our organization’s commitment to translating research innovations into operational capabilities that serve the broader water resources community.

Following the presentations, many researchers expressed immediate interest in using NextGen In A Box for both their research applications and operational forecasting. The tool's modular design, ease of deployment, and compatibility with diverse datasets resonated strongly with attendees seeking flexible and scalable modeling solutions. Several participants noted its potential for accelerating innovation in hydrologic modeling and expressed intent to integrate it into their ongoing and future projects.

The conference provided valuable opportunities for our researchers to connect with operational forecasters, emergency managers, and fellow scientists, fostering collaborations that will enhance the real-world impact of CIROH’s research initiatives. These interactions are essential for ensuring our scientific advances address the practical challenges faced by those responsible for flood warning and water resource management.

Looking Ahead​

CIROH’s strong presence at NHWC 2025 underscores our dedication to bridging the gap between research and operations in hydrologic science. Our researchers’ contributions to model evaluation methodologies and NextGen ecosystem development represent key steps toward more reliable, accessible, and effective water prediction systems.

As we continue to advance hydrologic science, conferences like NHWC 2025 provide essential platforms for sharing innovations, gathering feedback from the operational community, and building the collaborative relationships that drive progress in water resources management and flood warning systems.

The success of our team at NHWC 2025 reflects CIROH’s growing impact in the hydrologic sciences community and our commitment to developing solutions that serve both scientific advancement and societal needs.

My Experience at Tethys Summit 2025​

Earlier this month, I had the opportunity to attend Tethys Summit 2025 in Tampa, FL. It was a rewarding experience to learn about the Tethys Platform and how researchers, hydrologists, and geospatial scientists are applying it in their work. Through workshops and technical demonstrations, I gained insights into how this open-source Earth science platform is advancing environmental problem-solving.

On Day 1, I participated in the Tethys Dash workshop led by Corey Krewson, where I learned to build a Tethys dashboard without writing any code. During the hands-on session, I was surprised by how easily I was able to build dashboards using a drag-and-drop interface, as much of the technical implementation was abstracted for simplicity. I also attended the Tethys Component Apps (ReactPy) workshop by Shawn Crawley. Unlike the no-code approach of the first session, this workshop focused on building a basic interactive Tethys web app in under 50 lines of Python code.

On Day 2, I also took part in the Contributing to Tethys Core workshop led by Nathan Swain. This workshop outlined the steps for first-time contributors to get involved and submit pull requests to the Tethys Core repository. Aside from the workshops, I took part in Tethys App Competition where I presented the Tethys app I developed. It was a great learning experience to understand more about web app developments using Tethys. I ended up winning second place and got a bonsai tree LEGO set!

Overall, it was a very engaging and informative summit. For anyone who missed the session, the workshop materials are available here: Talks and Workshop Materials

— Manjila

What is Tethys?​

Tethys Platform is an open-source, Python-powered framework that lets geoscientists and developers turn complex environmental data and models into browser-based, interactive web GIS apps—without having to become web gurus. Built on Django, it bundles a full spatial stack (PostGIS, GeoServer, OpenLayers/Cesium) plus handy "Gizmos" for maps, charts, and dashboards, and it integrates seamlessly with scientific libraries like NumPy, Pandas, Plotly, and Bokeh. A built-in job manager runs heavy climate or hydrologic models—locally or in the cloud—while APIs make it easy to pull data from THREDDS servers or PostGIS databases. The result is a low-barrier path to custom interfaces, 2D/3D visualizations, and scalable computing that's already powering water-resource tools, climate-scenario explorers, and disaster-monitoring apps worldwide. Backed by the Tethys Geoscience Foundation, the platform's active open-source community keeps it evolving to meet new Earth-science challenges.

Community Impact​

Thanks to these capabilities, Tethys Platform is widely adopted in the geoscience community. Scientists around the world use it for a range of applications—water resource management, climate change analysis, natural disaster monitoring, environmental planning, and more. In each case, Tethys helps turn complex datasets and model outputs into accessible tools for decision-makers. As an open-source project under the Tethys Geoscience Foundation, it benefits from an active user community that continually contributes improvements and new features. This collaborative growth ensures the platform keeps pace with emerging needs in Earth science, effectively lowering the barrier for geoscientists to share data and models as powerful web applications.

Tethys Platform Resources​

Tethys Platform offers several resources to help you get started, including comprehensive official documentation, step-by-step tutorials, and a live demo portal. The official documentation covers everything from installation and "Hello World" examples to advanced topics and architecture (explaining how the Django backend and various services fit together). For hands-on learning, the tutorials guide you through setting up your first app, incorporating spatial data with built-in map layouts, connecting to GeoServer or THREDDS data servers, and even running cloud computations. And if you want to see Tethys in action without installing anything, the Tethys Demo Portal showcases a gallery of live example apps, such as a Gizmo Showcase of interactive widgets and a Wildfire Visualizer that displays wildfire data on a map, giving you a clear idea of the platform's capabilities and inspiring ideas for your own geoscientific web applications.

Each of these resources will help you dive deeper into the Tethys Platform. Whether you're looking to develop a water resources tool, a climate data viewer, or any geoscientific web application, the combination of documentation, tutorials, and live demos will support you on your journey with Tethys. Happy app building!

— Gio