# Source: JRC Global Surface Water
# Type: geotiff
# Catalog: Manual
# Created for: CBGB benchmark problems

source:
  id: jrc-global-surface-water
  name: JRC Global Surface Water v1.4
  type: geotiff
  url: https://global-surface-water.appspot.com/
  description: >
    The Global Surface Water dataset from the European Commission Joint Research
    Centre (JRC) maps the location and temporal distribution of surface water
    from 1984 to 2021 at 30m resolution. Based on analysis of over 4 million
    Landsat scenes, the dataset quantifies changes in global surface water over
    the past 38 years. Includes layers for water occurrence, change intensity,
    seasonality, recurrence, transitions, and maximum extent.
  provider: European Commission Joint Research Centre (JRC)
  license: CC-BY-4.0
  metadata:
    connector_id: null
    tags:
    - Water
    - Surface Water
    - Global
    - Landsat
    - 30m
    - JRC
    - Hydrology
    dataset:
      tier: 1
      thematic_group: Hydrology & Water
      format: geotiff
      resolution: 30m
      crs: EPSG:4326
      bands:
      - id: occurrence
        description: Frequency of water presence (0-100%)
        dtype: uint8
      - id: change
        description: Change in water occurrence between first and last period (-100 to 100)
        dtype: int8
      - id: seasonality
        description: Number of months water is present per year (1-12)
        dtype: uint8
      - id: recurrence
        description: Frequency of seasonal water return (0-100%)
        dtype: uint8
      - id: transitions
        description: Categorical change between water states (10 classes)
        dtype: uint8
      - id: max_extent
        description: Binary mask of maximum water extent 1984-2021
        dtype: uint8
      temporal:
        range: 1984-03-16/2021-12-31
        frequency: static
        type: snapshot
        default_range: latest
      quality:
        methodology: >
          Expert System classifier applied to Landsat 5 TM, 7 ETM+, and 8 OLI
          imagery. Each pixel classified as water/non-water per observation.
          Aggregated statistics computed over the full time series.
        accuracy: >
          Commission and omission errors <1% at pixel level for water detection.
          See Pekel et al. 2016, Nature 540(7633).
        coverage: global
      layout:
        type: tiled
        spatial:
          scheme: grid-10x10
        uri_template: https://storage.googleapis.com/global-surface-water/downloads2021/{band}/{lon}_{lat}.tif
        parameters:
        - var: band
          values:
          - occurrence
          - change
          - seasonality
          - recurrence
          - transitions
          - extent
      original_source_url: https://global-surface-water.appspot.com/download
      homepage: https://global-surface-water.appspot.com/
      gee_asset_id: JRC/GSW1_4/GlobalSurfaceWater
      cbgb_problems:
      - EBA_A1.5_A1
      - FRC_10
      - FRC_11
      - FRC_12