Add GEBCO, IBCSO, and IBCAO bathymetry dataset modules

Project.toml

@@ -33,6 +33,7 @@ Thermodynamics = "b60c26fb-14c3-4610-9d3e-2d17fe7ff00c"
 ZipFile = "a5390f91-8eb1-5f08-bee0-b1d1ffed6cea"
 
 [weakdeps]
+ArchGDAL = "c9ce4bd3-c3d5-55b8-8973-c0e20141b8c3"
 Breeze = "660aa2fb-d4c8-4359-a52c-9c057bc511da"
 CDSAPI = "8a7b9de3-9c00-473e-88b4-7eccd7ef2fea"
 CondaPkg = "992eb4ea-22a4-4c89-a5bb-47a3300528ab"
@@ -44,6 +45,7 @@ WorldOceanAtlasTools = "04f20302-f1b9-11e8-29d9-7d841cb0a64a"
 XESMF = "2e0b0046-e7a1-486f-88de-807ee8ffabe5"
 
 [extensions]
+NumericalEarthArchGDALExt = "ArchGDAL"
 NumericalEarthBreezeExt = "Breeze"
 NumericalEarthCDSAPIExt = "CDSAPI"
 NumericalEarthCopernicusMarineExt = "CopernicusMarine"
@@ -54,6 +56,8 @@ NumericalEarthWOAExt = "WorldOceanAtlasTools"
 
 [compat]
 Adapt = "4"
+ArchGDAL = "0.10"
+
 Breeze = "0.4"
 CDSAPI = "2.2.1"
 CFTime = "0.1, 0.2"

docs/src/Metadata/metadata_overview.md

@@ -80,12 +80,18 @@ NumericalEarth currently ships connectors for the following data products:
 
 | Dataset            | Supported Variables                                      | Documentation Link                                                                                 |
 |--------------------|-----------------------------------------------------------|-----------------------------------------------------------------------------------------------------|
+| **Bathymetry**     |                                                           |                                                                                                     |
 | `ETOPO2022`        | [Supported variables](@ref dataset-etopo2022-vars)        | [NOAA ETOPO 2022 overview](https://www.ncei.noaa.gov/products/etopo-global-relief-model)           |
+| `GEBCO2024`        | [Supported variables](@ref dataset-gebco2024-vars)        | [GEBCO 2024 overview](https://www.gebco.net/data_and_products/gridded_bathymetry_data/)             |
+| `IBCSOv2`          | [Supported variables](@ref dataset-ibcsov2-vars)          | [IBCSO overview](https://ibcso.org/ibcso-2024-annual-release/)                                      |
+| `IBCAOv5`          | [Supported variables](@ref dataset-ibcaov5-vars)          | [IBCAO overview](https://www.gebco.net/data_and_products/gridded_bathymetry_data/arctic_ocean/)     |
+| **Ocean reanalysis** |                                                         |                                                                                                     |
 | `ECCO2Monthly`     | [Supported variables](@ref dataset-ecco2monthly-vars)     | [ECCO2 documentation](https://ecco.jpl.nasa.gov/products/all/)                                     |
 | `ECCO2Daily`       | [Supported variables](@ref dataset-ecco2daily-vars)       | [ECCO2 documentation](https://ecco.jpl.nasa.gov/products/all/)                                     |
 | `ECCO4Monthly`     | [Supported variables](@ref dataset-ecco4monthly-vars)     | [ECCO V4r4 product guide](https://ecco-group.org/products-ECCO-V4r4.htm)                           |
 | `EN4Monthly`       | [Supported variables](@ref dataset-en4monthly-vars)       | [Met Office EN4 overview](https://www.metoffice.gov.uk/hadobs/en4/)                                |
 | `GLORYSDaily`      | [Supported variables](@ref dataset-glorysdaily-vars)      | [Copernicus GLORYS product page](https://data.marine.copernicus.eu/product/GLOBAL_MULTIYEAR_PHY_001_030/description) |
 | `GLORYSMonthly`    | [Supported variables](@ref dataset-glorysmonthly-vars)    | [Copernicus GLORYS product page](https://data.marine.copernicus.eu/product/GLOBAL_MULTIYEAR_PHY_001_030/description) |
+| **Atmospheric forcing** |                                                      |                                                                                                     |
 | `RepeatYearJRA55`  | [Supported variables](@ref dataset-repeatyearjra55-vars)  | [JRA-55 Reanalysis](https://www.data.jma.go.jp/jra/html/JRA-55/index_en.html)                                 |
 | `MultiYearJRA55`   | [Supported variables](@ref dataset-multiyearjra55-vars)   | [JRA-55 Reanalysis](https://www.data.jma.go.jp/jra/html/JRA-55/index_en.html)                                 |

docs/src/Metadata/supported_variables.md

@@ -4,13 +4,19 @@ NumericalEarth currently ships connectors for the following data products:
 
 | Dataset            | Supported Variables                                      | Documentation Link                                                                                 |
 |--------------------|-----------------------------------------------------------|-----------------------------------------------------------------------------------------------------|
+| **Bathymetry**     |                                                           |                                                                                                     |
 | `ETOPO2022`        | [Supported variables](@ref dataset-etopo2022-vars)        | [NOAA ETOPO 2022 overview](https://www.ncei.noaa.gov/products/etopo-global-relief-model)           |
+| `GEBCO2024`        | [Supported variables](@ref dataset-gebco2024-vars)        | [GEBCO 2024 overview](https://www.gebco.net/data_and_products/gridded_bathymetry_data/)             |
+| `IBCSOv2`          | [Supported variables](@ref dataset-ibcsov2-vars)          | [IBCSO overview](https://ibcso.org/ibcso-2024-annual-release/)                                      |
+| `IBCAOv5`          | [Supported variables](@ref dataset-ibcaov5-vars)          | [IBCAO overview](https://www.gebco.net/data_and_products/gridded_bathymetry_data/arctic_ocean/)     |
+| **Ocean reanalysis** |                                                         |                                                                                                     |
 | `ECCO2Monthly`     | [Supported variables](@ref dataset-ecco2monthly-vars)     | [ECCO2 documentation](https://ecco.jpl.nasa.gov/products/all/)                                     |
 | `ECCO2Daily`       | [Supported variables](@ref dataset-ecco2daily-vars)       | [ECCO2 documentation](https://ecco.jpl.nasa.gov/products/all/)                                     |
 | `ECCO4Monthly`     | [Supported variables](@ref dataset-ecco4monthly-vars)     | [ECCO V4r4 product guide](https://ecco-group.org/products-ECCO-V4r4.htm)                           |
 | `EN4Monthly`       | [Supported variables](@ref dataset-en4monthly-vars)       | [Met Office EN4 overview](https://www.metoffice.gov.uk/hadobs/en4/)                                |
 | `GLORYSDaily`      | [Supported variables](@ref dataset-glorysdaily-vars)      | [Copernicus GLORYS product page](https://data.marine.copernicus.eu/product/GLOBAL_MULTIYEAR_PHY_001_030/description) |
 | `GLORYSMonthly`    | [Supported variables](@ref dataset-glorysmonthly-vars)    | [Copernicus GLORYS product page](https://data.marine.copernicus.eu/product/GLOBAL_MULTIYEAR_PHY_001_030/description) |
+| **Atmospheric forcing** |                                                      |                                                                                                     |
 | `RepeatYearJRA55`  | [Supported variables](@ref dataset-repeatyearjra55-vars)  | [JRA-55 Reanalysis](https://www.data.jma.go.jp/jra/html/JRA-55/index_en.html)                                 |
 | `MultiYearJRA55`   | [Supported variables](@ref dataset-multiyearjra55-vars)   | [JRA-55 Reanalysis](https://www.data.jma.go.jp/jra/html/JRA-55/index_en.html)                                 |
 
@@ -114,3 +120,15 @@ NumericalEarth currently ships connectors for the following data products:
 - `:snow_freshwater_flux` - Precipitation flux from snow/ice (kg m⁻² s⁻¹).
 - `:river_freshwater_flux` - River discharge flux (kg m⁻² s⁻¹).
 - `:iceberg_freshwater_flux` - Iceberg calving flux (kg m⁻² s⁻¹).
+
+## [Supported variables for IBCSOv2](@id dataset-ibcsov2-vars)
+
+- `:bottom_height` - Southern Ocean bathymetry at 500 m resolution, south of 50°S (m).
+
+## [Supported variables for GEBCO2024](@id dataset-gebco2024-vars)
+
+- `:bottom_height` - Global bathymetry and topography at 15 arc-second resolution (m).
+
+## [Supported variables for IBCAOv5](@id dataset-ibcaov5-vars)
+
+- `:bottom_height` - Arctic Ocean bathymetry at 100 m resolution, north of 64°N, including Greenland ice sheet surface elevation (m).

ext/NumericalEarthArchGDALExt.jl

@@ -0,0 +1,50 @@
+module NumericalEarthArchGDALExt
+
+using NumericalEarth
+using ArchGDAL
+using NCDatasets
+
+import NumericalEarth.DataWrangling.IBCAO: reproject_ibcao_to_netcdf
+
+function reproject_ibcao_to_netcdf(tiff_path, nc_path)
+    ArchGDAL.read(tiff_path) do src
+        # Warp from EPSG:3996 (Polar Stereographic) to EPSG:4326 (WGS84)
+        # at 0.01° resolution, clipping to 64–90°N
+        ArchGDAL.gdalwarp([src],
+            ["-t_srs", "EPSG:4326",
+             "-te",    "-180", "64", "180", "90",  # xmin ymin xmax ymax
+             "-tr",    "0.01", "0.01",             # target resolution (degrees)
+             "-r",     "bilinear",                 # resampling method
+             "-ot",    "Float32"]) do warped
+
+            # ArchGDAL returns data as (Nx, Ny) with y from north to south (GDAL convention)
+            data = Float32.(ArchGDAL.read(warped, 1))
+            data = data[:, end:-1:1]  # flip y: j=1 → 64°N (south), j=Ny → 90°N (north)
+
+            Nx, Ny = size(data)  # expected: (36000, 2600)
+
+            NCDataset(nc_path, "c") do ds
+                defDim(ds, "lon", Nx)
+                defDim(ds, "lat", Ny)
+
+                lon_var = defVar(ds, "lon", Float64, ("lon",);
+                                attrib = ["units" => "degrees_east",
+                                          "long_name" => "longitude"])
+                lat_var = defVar(ds, "lat", Float64, ("lat",);
+                                attrib = ["units" => "degrees_north",
+                                          "long_name" => "latitude"])
+                z_var   = defVar(ds, "z",   Float32, ("lon", "lat");
+                                attrib = ["long_name" => "elevation",
+                                          "units"     => "m"])
+
+                lon_var[:] = range(-180 + 0.005, 180 - 0.005; length=Nx)
+                lat_var[:] = range(64 + 0.005, 90 - 0.005; length=Ny)
+                z_var[:, :] = data
+            end
+        end
+    end
+
+    return nothing
+end
+
+end # module NumericalEarthArchGDALExt

src/Bathymetry/Bathymetry.jl

@@ -19,7 +19,8 @@ using NCDatasets
 using Printf
 using Scratch
 
-using ..DataWrangling: Metadatum, native_grid, metadata_path, download_dataset
+using ..DataWrangling: Metadatum, native_grid, metadata_path, download_dataset,
+                      dataset_variable_name, validate_dataset_coverage
 using ..DataWrangling.ETOPO: ETOPO2022
 
 include("regrid_bathymetry.jl")

src/Bathymetry/regrid_bathymetry.jl

@@ -186,6 +186,8 @@ function regrid_bathymetry(target_grid, metadata;
                            major_basins = 1,
                            cache = true)
 
+    validate_dataset_coverage(target_grid, metadata)
+
     config = BathymetryRegridding(target_grid, metadata;
                                   height_above_water, minimum_depth,
                                   interpolation_passes, major_basins)
@@ -240,7 +242,7 @@ function _regrid_bathymetry(target_grid, metadata;
     filepath = metadata_path(metadata)
     dataset = Dataset(filepath, "r")
 
-    z_data = convert(Array{FT}, dataset["z"][:, :])
+    z_data = convert(Array{FT}, dataset[dataset_variable_name(metadata)][:, :])
     close(dataset)
 
     if !isnothing(height_above_water)

src/DataWrangling/DataWrangling.jl

@@ -5,6 +5,7 @@ restoring, or validation.
 module DataWrangling
 
 export Metadata, Metadatum, DatewiseFilename, ECCOMetadatum, EN4Metadatum, all_dates, first_date, last_date
+export validate_dataset_coverage, metadata_filename
 export BoundingBox, Column, Linear, Nearest
 export WOAClimatology, WOAAnnual, WOAMonthly
 export metadata_time_step, metadata_epoch
@@ -198,6 +199,29 @@ function binary_data_size end
 
 default_mask_value(dataset) = NaN
 
+"""
+    AbstractStaticDataset
+
+Supertype for datasets without a time dimension. Provides default no-op implementations for the date-related interface 
+(`all_dates`, `first_date`, `last_date`).
+"""
+abstract type AbstractStaticDataset end
+
+all_dates(::AbstractStaticDataset,  args...) = nothing
+first_date(::AbstractStaticDataset, args...) = nothing
+last_date(::AbstractStaticDataset,  args...) = nothing
+
+"""
+    AbstractStaticBathymetry <: AbstractStaticDataset
+
+Supertype for static, two-dimensional bathymetry datasets (e.g. ETOPO, GEBCO, IBCSO, IBCAO). 
+Adds defaults for the degenerate vertical axis and a variable-agnostic `Base.size`.
+"""
+abstract type AbstractStaticBathymetry <: AbstractStaticDataset end
+
+z_interfaces(::AbstractStaticBathymetry) = (0, 1)
+Base.size(dataset::AbstractStaticBathymetry, variable) = size(dataset)
+
 # Fundamentals
 include("metadata.jl")
 include("metadata_field.jl")
@@ -231,6 +255,9 @@ include("ORCA/ORCA.jl")
 include("WOA/WOA.jl")
 include("JRA55/JRA55.jl")
 include("OSPapa/OSPapa.jl")
+include("IBCSO/IBCSO.jl")
+include("GEBCO/GEBCO.jl")
+include("IBCAO/IBCAO.jl")
 
 using .ETOPO
 using .ECCO
@@ -241,6 +268,9 @@ using .ORCA
 using .WOA
 using .JRA55
 using .OSPapa
+using .IBCSO
+using .GEBCO
+using .IBCAO
 
 # Fallback: if no download extension is loaded, check that all files already exist
 function download_dataset(metadata::Metadata)

src/DataWrangling/ETOPO/ETOPO.jl

@@ -7,19 +7,15 @@ using Oceananigans
 using Oceananigans.DistributedComputations: @root
 using Scratch
 
-using ..DataWrangling: download_progress, Metadatum, metadata_path
+using ..DataWrangling: download_progress, Metadatum, metadata_path, AbstractStaticBathymetry
 
 import NumericalEarth.DataWrangling:
     metadata_filename,
     default_download_directory,
-    all_dates,
-    first_date,
-    last_date,
     dataset_variable_name,
     download_dataset,
     longitude_interfaces,
     latitude_interfaces,
-    z_interfaces,
     reversed_vertical_axis
 
 download_ETOPO_cache::String = ""
@@ -31,18 +27,13 @@ ETOPO_bathymetry_variable_names = Dict(
     :bottom_height => "z",
 )
 
-struct ETOPO2022 end
+struct ETOPO2022 <: AbstractStaticBathymetry end
 
 default_download_directory(::ETOPO2022) = download_ETOPO_cache
 reversed_vertical_axis(::ETOPO2022) = true
 longitude_interfaces(::ETOPO2022) = (-180, 180)
 latitude_interfaces(::ETOPO2022) = (-90, 90)
 Base.size(::ETOPO2022) = (21600, 10800, 1)
-Base.size(dataset::ETOPO2022, variable) = size(dataset)
-
-all_dates(::ETOPO2022, args...) = nothing
-first_date(::ETOPO2022, args...) = nothing
-last_date(::ETOPO2022, args...) = nothing
 
 const ETOPOMetadatum = Metadatum{<:ETOPO2022}
 
@@ -51,7 +42,6 @@ dataset_variable_name(data::ETOPOMetadatum) = ETOPO_bathymetry_variable_names[da
 const ETOPO_url = "https://www.dropbox.com/scl/fi/6pwalcuuzgtpanysn4h6f/" *
     "ETOPO_2022_v1_60s_N90W180_surface.nc?rlkey=2t7890ruyk4nd5t5eov5768lt&st=yfxsy1lu&dl=0"
 
-z_interfaces(::ETOPOMetadatum) = (0, 1)
 metadata_url(::ETOPOMetadatum) = ETOPO_url
 metadata_filename(::ETOPO2022, name, date, region) = "ETOPO_2022_v1_60s_N90W180_surface.nc"
 

src/DataWrangling/GEBCO/GEBCO.jl

@@ -0,0 +1,125 @@
+module GEBCO
+
+export GEBCO2024
+
+using Downloads
+using ZipFile
+using Oceananigans
+using Oceananigans.DistributedComputations: @root
+using Scratch
+
+using ..DataWrangling: download_progress, Metadatum, metadata_path, AbstractStaticBathymetry
+
+import NumericalEarth.DataWrangling:
+    metadata_filename,
+    default_download_directory,
+    dataset_variable_name,
+    download_dataset,
+    longitude_interfaces,
+    latitude_interfaces,
+    reversed_vertical_axis
+
+download_GEBCO_cache::String = ""
+function __init__()
+    global download_GEBCO_cache = @get_scratch!("GEBCO")
+end
+
+GEBCO_bathymetry_variable_names = Dict(
+    :bottom_height => "elevation",  # Variable name in NetCDF
+)
+
+"""
+    GEBCO2024
+
+General Bathymetric Chart of the Oceans 2024 release.
+Global bathymetry and topography at 15 arc-second resolution.
+
+The GEBCO_2024 Grid is a global terrain model for ocean and land,
+providing elevation data on a 15 arc-second interval grid.
+
+Reference: GEBCO Compilation Group (2024) GEBCO 2024 Grid
+Data source: https://www.gebco.net/data_and_products/gridded_bathymetry_data/
+"""
+struct GEBCO2024 <: AbstractStaticBathymetry end
+
+default_download_directory(::GEBCO2024) = download_GEBCO_cache
+reversed_vertical_axis(::GEBCO2024) = false
+
+# GEBCO covers the entire globe
+longitude_interfaces(::GEBCO2024) = (-180, 180)
+latitude_interfaces(::GEBCO2024) = (-90, 90)
+
+# Grid size for 15 arc-second resolution
+# 360° / (15/3600)° = 86400 points in longitude
+# 180° / (15/3600)° = 43200 points in latitude
+Base.size(::GEBCO2024) = (86400, 43200, 1)
+
+const GEBCOMetadatum = Metadatum{<:GEBCO2024}
+
+dataset_variable_name(data::GEBCOMetadatum) = GEBCO_bathymetry_variable_names[data.name]
+
+# GEBCO 2024 download URL from BODC
+# Note: This is a large file (~8 GB zipped, ~22 GB unzipped)
+const GEBCO_zip_url = "https://www.bodc.ac.uk/data/open_download/gebco/gebco_2024/zip/"
+
+# The expected NetCDF filename inside the ZIP
+const GEBCO_nc_filename = "GEBCO_2024.nc"
+metadata_filename(::GEBCO2024, name, date, bounding_box) = GEBCO_nc_filename
+
+function download_dataset(metadatum::GEBCOMetadatum)
+    filepath = metadata_path(metadatum)
+    download_dir = metadatum.dir
+
+    @root if !isfile(filepath)
+        @info "Downloading GEBCO data: $(metadatum.name) to $download_dir..."
+        @info "Note: GEBCO is a large dataset (~8 GB download, ~22 GB uncompressed). This may take a while."
+
+        # Download the ZIP file
+        zip_path = joinpath(download_dir, "GEBCO_2024.zip")
+
+        try
+            @info "Downloading from BODC..."
+            Downloads.download(GEBCO_zip_url, zip_path; progress=download_progress)
+
+            # Extract the NetCDF file from the ZIP using ZipFile.jl
+            @info "Extracting NetCDF from ZIP archive..."
+            zf = ZipFile.Reader(zip_path)
+            extracted = false
+            for f in zf.files
+                if endswith(f.name, GEBCO_nc_filename)
+                    open(filepath, "w") do io
+                        write(io, read(f))
+                    end
+                    extracted = true
+                    break
+                end
+            end
+            close(zf)
+
+            if !extracted
+                error("Could not find $GEBCO_nc_filename in ZIP archive")
+            end
+
+            if isfile(filepath)
+                @info "GEBCO data extracted successfully"
+            else
+                error("Failed to extract GEBCO NetCDF file")
+            end
+
+            # Clean up ZIP file to save space
+            rm(zip_path; force=true)
+
+        catch e
+            @warn "Failed to download GEBCO: $e"
+
+            # Clean up any partial download
+            rm(zip_path; force=true)
+
+            rethrow(e)
+        end
+    end
+
+    return filepath
+end
+
+end # module