United States Reservoir Transparency

The United States operates the deepest and most pluralistic public reservoir data ecosystem in the world. USGS NWIS publishes daily storage (parameter code 00054) and stage readings for thousands of lake/reservoir sites. Bureau of Reclamation publishes daily storage for its ~190 reservoirs across the 17 western states via the RISE platform and the Hydrologic Database (HDB). USACE publishes daily storage for ~700 water-control reservoirs across 16 districts via the CWMS Data API. California DWR's CDEC publishes near-real-time sub-daily storage for the entire state. TVA publishes daily levels for its 49 Tennessee Valley reservoirs. Texas Water Development Board publishes daily readings for all major Texas reservoirs. Combined, these systems make daily volumetric storage publicly available for a clear majority of nationally relevant reservoirs (>10 hm³).

USGS Water Services is a formally documented, versioned, anonymous-access REST API supporting JSON, WaterML and RDB formats — among the best-engineered public hydrological APIs in the world. BOR RISE exposes a JSON:API specification with full Swagger/OpenAPI documentation at data.usbr.gov/rise/api/swagger. USACE CWMS Data API (cwms-data.usace.army.mil) returns JSON time series for all its reservoirs without authentication. CDEC provides both CSV download endpoints and a JSON API. All federal data is released into the public domain (USC Title 17 §105). The only friction — and the reason this is 92 rather than 100 — is institutional fragmentation: integrating these systems requires reconciling different parameter codes, station identifiers, unit conventions (acre-feet vs cubic meters vs cubic feet) and timestamp schemes, with no single national endpoint. This was directly observed building reservoirs.earth's 559-reservoir US dataset: every major federal API worked anonymously and reliably, but stitching them together required a multi-source builder.

Methodology v1.3.0 (capacity-weighted). Total US national reservoir storage capacity per the National Inventory of Dams (NID) is ~830,000 hm³ (the operationally relevant >10 hm³ subset accounting for essentially all national volume). Covered capacity, summing the federal and state public daily feeds — USGS NWIS reservoir/lake gauges (~150,000 hm³ across thousands of sites), Bureau of Reclamation (~120,000 hm³ for ~190 reservoirs including the giants Lake Mead 32,000 hm³ and Lake Powell 30,000 hm³), USACE CWMS (~130,000 hm³ across ~700 reservoirs), CDEC California (~50,000 hm³ across ~150 reservoirs including Shasta 5,615 hm³ and Oroville 4,364 hm³), TVA (~30,000 hm³ across 49 reservoirs), TWDB Texas (~30,000 hm³), plus state agencies in Pacific Northwest, Mountain West, Florida, Georgia and the Carolinas (~30,000 hm³ additional, net of overlap) — totals approximately 480,000 hm³ of covered storage. Coverage = round(100 × 480,000 / 830,000) = 58. The capacity-weighted result is similar to the prior count-based number because USA's largest reservoirs (Mead, Powell, Sakakawea, Oahe, Fort Peck, Garrison) are all federally operated and publicly tracked, but a substantial volume sits in non-federal hydropower (TVA's neighbors in the Southeast, some private utility dams) and in municipal water-supply systems in the Northeast where no machine-readable storage feed exists.

USGS NWIS provides continuous daily records reaching back to the 1900s for the oldest gauges, with most major reservoir gauges having unbroken daily series from the 1930s–1970s onward. BOR's records for Lake Mead and Lake Powell span the entire history of those reservoirs (since 1935 and 1963 respectively). USACE CWMS holds decades of daily records for its reservoirs. CDEC archives California reservoir data back to the 1980s for many sites and to the 1960s for others. The combination of 90+ years of continuous daily federal records, fully accessible through public APIs, is unmatched globally — exceeding Spain's 35-year weekly series and Norway's 30-year weekly series. This depth enables climate trend analysis (PDSI reconstructions, megadrought studies) that no other country's reservoir data infrastructure supports.

USGS NWIS publishes daily mean values with same-day or next-day availability and 15-minute instantaneous values for many sites. BOR RISE updates daily. USACE CWMS publishes near-real-time hourly data for many reservoirs. CDEC publishes sub-daily readings (hourly or better) for California reservoirs. TVA updates lake levels multiple times daily. The aggregate update cadence is effectively real-time or near-real-time for the bulk of covered reservoirs — only the absence of a single unified national daily aggregate (comparable to Brazil's ONS three-times-daily national CSV) prevents a higher score.

USGS publishes extensive documentation: the National Field Manual for Water-Quality Data, parameter code dictionaries (over 25,000 codes), data quality flags (Approved, Provisional, Estimated), site metadata (datum, elevation, drainage area, operator) and detailed measurement methodology. BOR RISE documents data provenance and source agency for every series. USACE publishes CWMS data standards. CDEC publishes sensor metadata and quality codes. Reservoir capacity (the denominator needed to compute % full) is documented in the National Inventory of Dams and in operator-published rating curves, though it is not always co-located with the storage time series and requires cross-referencing. Cross-agency consistency is the principal gap: USGS, BOR and USACE all use different station identifiers and parameter dictionaries, and harmonising them is left to the user.

All documentation, APIs, portals and metadata are in English — the global lingua franca for scientific data. The USGS, BOR, USACE, CDEC and TVA web portals are modern, mobile-responsive applications with interactive data explorers, mapping interfaces and chart tools. Federal data is released into the public domain with no licensing barriers for international users. This is the easiest reservoir data in the world to consume from any country.