Resources
Pages for the work at hand: start a model, judge its quality, use the data, connect through the API, and ask us for a study.
Start a model
Watershed Explorer Explore data and layers, build a SWAT+ model, and analyze charts and reports.SWAT+ model generation Build a SWAT+ model. Get a package you can open and run.How it works Pick a USGS gage or HUC12, confirm settings, track the cloud queue, download Editor-ready packages.Example models Browse and download completed SWAT+ packages built from NHDPlus HR, 3DEP, gSSURGO, NLCD, and PRISM.Documentation Model bundle structure, SWAT+ input file deltas, PFAS fate-and-transport files, engine run modes, and MODFLOW coupling.Films Short films about watershed modeling and the platform, each with its full transcript.
Judge the quality
How we check models Every check we run on a SWAT+ model, in order: what it tests, its threshold, and whether a failure stops the build, a release or a calibration.Methodology Catchment prep, HRUs, climate inputs, QSWAT+ assembly, QA checks, and downloadable Editor packages.Calibration methods SWATGenX calibrates SWAT+ hydrology against USGS streamflow with MMPSO, a multi-memory, role-aware particle swarm.Calibration workspace Pick a model, review its observation coverage, choose parameters, and get a wall-time estimate.Matching USGS gages to SWAT+ channels How SWATGenX assigns USGS streamgages to SWAT+ routed channels using NHDPlus HR, NWIS metadata, and lake and canal context.Drainage area audit Drainage-area closure audit: SWAT+ areas against NHDPlus HR and NWIS at stream gages.
Use the data
Groundwater inventory A national database of water-well lithology and driller logs, with water levels, yields and hydraulic conductivity.Where PFAS data is public Ambient PFAS monitoring is broadly public; depth-resolved soil reports mostly are not. What exists, where it lives, and what is missing.National PFAS soil inventory A cited Zenodo deposit of soil PFAS records; its data files are not public. Browse the inventory on SWATGenX.National PFAS analysis From a national PFAS inventory to watershed predictions.Climate change and streamflow Downscaled CMIP6 climate models run through one SWAT+ model of a Michigan river: temperature, ET, snow, timing and flow, with ranges.
Connect through the API
Developers Three programmatic doors into SWATGenX: the swatgenx Python package on PyPI, a REST model-generation API, and a public MCP server for AI agents.Python package The swatgenx PyPI package: public data functions that need no account, and authenticated build requests and downloads through Client.MCP server A public Model Context Protocol server over streamable HTTP, exposing watershed modelling and national hydrologic datasets to AI agents.I/O reference File-level reference for what a SWATGenX package contains and how the SWATGenX engine extensions read and write it.
Studies and services
Learn watershed modeling. A growing curriculum on how water moves and how it is modelled, from the water cycle to groundwater.Education