Add z_top to clm artifacts and rename them

Adds Canopy height to both clm data artifacts.
Also rename the low res artifact from clm_data
to clm_data_0.9x1.25 and the high res artifact from
clm_data_0.125 to clm_data_0.125x0.125

clm_data/OutputArtifacts.toml

@@ -1,12 +1,12 @@
-[clm_data]
-git-tree-sha1 = "3f6873a3e67722bda1fd23f7d5a05f5e2df1fe8c"
+["clm_data_0.9x1.25"]
+git-tree-sha1 = "283c62220fca8c4afe36a62348d3e9a159af2ee9"
 
-    [[clm_data.download]]
-    sha256 = "830136abec15551a343b7884d657fb6a79b16a37237681a2becf65bd845aa692"
-    url = "https://caltech.box.com/shared/static/6pu2f6c99g29qawvjjh3j56drkonpd3z.gz"
-[clm_data_highres]
-git-tree-sha1 = "e68ba26e07c135f66d1ad678795b4a25c929b1b6"
+    [["clm_data_0.9x1.25".download]]
+    sha256 = "88d5899a729de800017a74bd8a7278582c2c55c0d8730a529bae384425d70e4e"
+    url = "https://caltech.box.com/shared/static/mxs3l1c1dppjwy81assk8w8ofn9d70pu.gz"
+["clm_data_0.125x0.125"]
+git-tree-sha1 = "6284ddefbc7937d9c1fb68fa731ff3f00b68e917"
 
-    [[clm_data_highres.download]]
-    sha256 = "2c7e6ba3c11e1e6ef209c7437aa18fb4cf15edd11332527bf0b41495dccbed19"
-    url = "https://caltech.box.com/shared/static/kk72iah76p99yggm8knu6me8fh9eiid0.gz"
+    [["clm_data_0.125x0.125".download]]
+    sha256 = "8d2a61baad53346515f4a66c4342f8aafce37ca9520020c17f99cc4f4aa98b15"
+    url = "https://caltech.box.com/shared/static/uteaw1l6fg7wmwhb3ey1c0jq4r8vg0ts.gz"

clm_data/README.md

@@ -19,8 +19,8 @@ To recreate the artifact:
 
 ### 1. `surfdata_0.9x1.25_16pfts__CMIP6_simyr2000_c170616.nc` and `surfdata_0.125x0.125_16pfts_simyr2000_c151014.nc`
 These netCDF files includes comprehensive environmental data with a focus on vegetation represented through different Plant Functional Types (PFTs). These PFTs play a crucial role in modeling biophysical processes and ecosystem functions within CLM simulations. It also contains soil color data, which is used to calculate soil alebdo. `surfdata_0.9x1.25_16pfts__CMIP6_simyr2000_c170616.nc` is on a 0.8x1.25 degree grid and is used to create the
-clm_data artifact, while `surfdata_0.125x0.125_16pfts_simyr2000_c151014.nc` is on a 0.125x0.125 degree
-grid and is used to create the clm_data_highres artifact. Both input files contain tge following PFTs:
+clm_data_0.9x1.25 artifact, while `surfdata_0.125x0.125_16pfts_simyr2000_c151014.nc` is on a 0.125x0.125 degree
+grid and is used to create the clm_data_0.125x0.125 artifact. Both input files contain the following PFTs:
 
 - **Plant Functional Types**:
   - `not_vegetated`
@@ -115,6 +115,9 @@ value of 0.0 if C4 is dominant.
   - `rooting_depth(lat, lon)` parameter for root_distribution (m)
     - Describes the depth where ~2/3 of the roots are above
   - `xl(lat, lon)` Leaf/stem orientation index
+  - `z_top(lat, lon)` height top (m) **NOTE**: This variable is always calculated using the
+  0.9x1.25 surface data file, because the `MONTHLY_HEIGHT_TOP` variable is
+  zero at every point in the 0.125x0.125 surface data file.
 
 ### 3. `soil_properties_map.nc`
 Contains the mapped soil albedos for each grid cell.

clm_data/create_artifact.jl

@@ -17,8 +17,8 @@ const FILE_PATHS = [
 const OUTPUT_FILES =
     ["dominant_PFT_map.nc", "vegetation_properties_map.nc", "soil_properties_map.nc"]
 
-output_dir = basename(@__DIR__) * "_artifact"
-output_dir_highres = basename(@__DIR__) * "_highres_artifact"
+output_dir = basename(@__DIR__) * "_0.9x1.25_artifact"
+output_dir_highres = basename(@__DIR__) * "_0.125x0.125_artifact"
 for dir in [output_dir, output_dir_highres]
     if isdir(dir)
         @warn "$dir already exists. Content will end up in the artifact and may be overwritten."
@@ -67,10 +67,10 @@ for output_file in OUTPUT_FILES
     Base.mv(output_file, output_path; force = true)
 end
 
-create_artifact_guided(output_dir; artifact_name = basename(@__DIR__))
+create_artifact_guided(output_dir; artifact_name = basename(@__DIR__) * "_0.9x1.25")
 
 create_artifact_guided(
     output_dir_highres;
-    artifact_name = basename(@__DIR__) * "_highres",
+    artifact_name = basename(@__DIR__) * "_0.125x0.125",
     append = true,
 )

clm_data/pft_variables.py

@@ -31,6 +31,14 @@
 # Get photosynthesis mechanisms from surface data file
 surface_dataset = nc.Dataset(surface_file, 'r')
 pft_values = surface_dataset.variables['PCT_NAT_PFT'][:]
+# the 0.125x0.125 file has all `MONTHLY_HEIGHT_TOP` values of zero everywhere, so we use the 0.9x1.25 file
+if args.detailed:
+    height_dataset = nc.Dataset('surfdata_0.9x1.25_16pfts__CMIP6_simyr2000_c170616.nc', 'r')
+    height_values = height_dataset.variables['MONTHLY_HEIGHT_TOP'][:, :, :, :]
+    height_lon = np.mean(height_dataset.variables['LONGXY'][:, :], axis=0)
+    height_lat = np.mean(height_dataset.variables['LATIXY'][:, :], axis=1)
+else:
+    height_values = surface_dataset.variables['MONTHLY_HEIGHT_TOP'][:, :, :, :]
 pft_dominant_values = np.argmax(pft_values, axis = 0)
 # pft index 14 is the only C4 pft
 c3_dominant_map = pft_dominant_values != 14
@@ -49,6 +57,7 @@
 xl_values = pft_physiology_dataset.variables['xl'][:]
 
 # Create arrays to store the mapped values for each grid point
+canopy_height_map = np.zeros(dominant_pft.shape, dtype=np.float32)
 medlynslope_map = np.zeros_like(dominant_pft, dtype=np.float32)
 medlynintercept_map = np.zeros_like(dominant_pft, dtype=np.float32)
 rholnir_map = np.zeros_like(dominant_pft, dtype=np.float32)
@@ -66,6 +75,16 @@
     for j in range(dominant_pft.shape[1]):
         pft_index = dominant_pft[i, j]  # PFTs are directly indexed
         if pft_index >= 0:
+            if args.detailed:
+                # here we use the lower resolution height data to map the canopy height
+                lon_point = lon[i, j]
+                lat_point = lat[i, j]
+                canopy_height_lon_index = np.argmin(np.abs(height_lon - lon_point))
+                canopy_height_lat_index = np.argmin(np.abs(height_lat - lat_point))
+                canopy_height_map[i,j] = np.mean(height_values[:, pft_index,\
+                    canopy_height_lat_index, canopy_height_lon_index], axis = 0)
+            else:
+                canopy_height_map[i,j] = np.mean(height_values[:, pft_index, i, j])
             medlynslope_map[i, j] = medlynslope_values[pft_index]
             medlynintercept_map[i, j] = medlynintercept_values[pft_index]
             # convert beta parameter to rooting depth parameter for root probability distribution
@@ -88,6 +107,7 @@
     # Create variables
     latitudes = output_dataset.createVariable('lat', 'f4', ('lat',))
     longitudes = output_dataset.createVariable('lon', 'f4', ('lon',))
+    canopy_height_var = output_dataset.createVariable('z_top', 'f4', ('lat', 'lon'), fill_value=np.nan)
     c3_dominant_var = output_dataset.createVariable('c3_dominant', 'f4', ('lat', 'lon',), fill_value=np.nan)
     medlynslope_var = output_dataset.createVariable('medlynslope', 'f4', ('lat', 'lon',), fill_value=np.nan)
     medlynintercept_var = output_dataset.createVariable('medlynintercept', 'f4', ('lat', 'lon',), fill_value=np.nan)
@@ -105,6 +125,7 @@
     # Assign data to variables
     latitudes[:] = np.mean(lat, axis=1)  # Assuming LATIXY and LONGXY are 2D arrays
     longitudes[:] = np.mean(lon, axis=0)  # Averaging to get 1D lat/lon
+    canopy_height_var[:, :] = canopy_height_map
     c3_dominant_var[:, :]  = c3_dominant_map
     medlynslope_var[:, :] = medlynslope_map
     medlynintercept_var[:, :] = medlynintercept_map
@@ -122,6 +143,9 @@
     # Assign attributes
     latitudes.units = 'degrees_north'
     longitudes.units = 'degrees_east'
+    canopy_height_var.units = 'm'
+    canopy_height_var.long_name = 'Canopy top height'
+    canopy_height_var.short_name = 'z_top'
     c3_dominant_var.units = '0. = c4, 1. = c3'
     c3_dominant_var.long_name = 'c3 dominant'
     medlynslope_var.units = 'kPa^0.5'