HydroAI Data

Datasets

Data products and observation networks distributed by the HydroAI Lab. Select a dataset to explore.

LiveCal/val site

NASA-GIST Core Validation Site

South Korea’s first Core Validation Site (CVS), at Hampyeong: one 1 km grid cell measured from the ground, from drones and from satellites, to check the soil moisture maps of NASA SMAP and NASA-ISRO NISAR.

Site
1 km × 1 km cell, six kinds of land cover
Ground
47 stations, 82 soil moisture and temperature sensors, every 5 minutes since June 2025
Drones
1 m maps every one to two weeks in season: true colour, NDVI, thermal, LiDAR, L-band radiometer
Satellites
HLS, Landsat, MODIS, Sentinel-1, NISAR (beta)
  • In the portal
  • Ground, drone and satellite views side by side, date by date
  • Each station’s series and its soil properties
  • Data files on request
Open portal →
LiveEvapotranspiration

Global 30m-ET

Prithvi-OpenET: monthly evapotranspiration at 30 m from Landsat and Sentinel-2 imagery, estimated with a geospatial foundation model. The Korean Peninsula and France are open now as a preview; other regions are produced on request.

Resolution
30 m, monthly
Record
April 2013 – August 2026, 160 monthly maps per region
Coverage
Korean Peninsula and France now; any other land area on request
Status
Paper under review. Until it is accepted the maps are shown down to about 120 m, and at 30 m in 19 sample areas
  • In the portal
  • Mean annual ET, the seasonal cycle, or any single month
  • Click a place for its monthly series and full record
  • Request a region
Open portal →
LiveSoil moisture

Hybrid Dual-Channel Algorithm (HDCA)

L-band soil moisture that retrieves the vegetation scattering albedo instead of assuming it: HDCA solves soil moisture, VOD and a time-varying effective scattering albedo together from SMAP brightness temperatures, with first-order canopy scattering, so dense forests no longer read as persistently wet. Over the Amazon basin the record follows the wet and dry seasons, as the picture shows.

Grid
36 km EASE-Grid 2.0, global land
Record
Daily, April 2015 – December 2025
Variables
Soil moisture, vegetation optical depth (VOD), effective scattering albedo
Status
Published in Remote Sensing of Environment (349, 115787); data open on Zenodo (CC BY 4.0)
  • In the portal
  • Daily, monthly and climatological maps on a globe
  • Click a cell for its full daily record of all three variables
  • Five validation sites marked
  • Zenodo files, citation and collaboration contact
Open portal →
LiveSoil moisture error

SMAP Time-Variant Error

How wrong is each SMAP soil-moisture value? The random error of every retrieval from triple collocation with ASCAT and GLDAS in a moving window, and, where triple collocation fails, predicted from SMAP’s own variables by LGBM and DNN models, with flags that say which is which and why.

Grid
36 km EASE-Grid 2.0, global land
Record
Every overpass, 6 a.m. and 6 p.m., February 2015 – November 2023
Variables
Soil moisture, TCA error σ, σ gap-filled by LGBM and by DNN, four TCA flags
Status
Published in GIScience & Remote Sensing (2025); 2022 on Zenodo, other years on request
  • In the portal
  • Overpass, monthly and climatological maps on a globe; where the error could be computed, filled, or not
  • Click a cell for its nine-year record with the four flags
  • Zenodo link, citation and collaboration contact
Open portal →
LiveWater quality

Inland Water Quality of South Korea

Six water quality parameters for the rivers and reservoirs of South Korea on a 1 km grid, each value with a calibrated uncertainty: water temperature, suspended solids, chlorophyll-a, dissolved oxygen, total nitrogen and total phosphorus, estimated from Sentinel-2 imagery, weather reanalysis and national monitoring records. The portal says what each layer can and cannot be used for.

Grid
1 km, inland waters of mainland South Korea
Record
1,012 clear overpass dates, January 2016 – December 2024
Variables
Six parameters, each with its uncertainty
Status
Data open on Zenodo (CC BY 4.0); paper under review at Scientific Data
  • In the portal
  • One date, a month, the seasonal cycle or the record mean, for each parameter and its uncertainty
  • Click a cell for every overpass at it, with the 95 % interval
  • Download links, code examples, and how far to trust each layer
Open portal →