Output Structure

Where everything lands and what each file contains.

The output tree is organised by correction stage and by purpose. Every NetCDF carries CF-1.8 attributes - use ncdump -h <file> or xr.open_dataset(<file>).attrs to inspect.

Top-level layout

data/output/                                      (or data/example_bali/output/ for Bali)
  corrected_ls/
  corrected_lseqm/
  corrected_lseqmdl/
  trained_models/
  station_density/
  metrics_ls/
  metrics_lseqm/
  metrics_lseqmdl/
  quality_ls/
  quality_lseqm/
  quality_lseqmdl/
  station_validation/
  figures/

Corrected precipitation NetCDFs

File pattern:

{prefix}_{method_abbr}_corrected_imergl_month{MM}_dekad{D}.nc4

Variable	Type	Units	Description
`precipitation`	float32	mm/day	Corrected daily precipitation
`lat`, `lon`	float64	deg	Cell-centre coordinates, 0.1 deg grid
`time`	datetime64	UTC	Daily timestamps for the dekad x all years

Global attributes record the method (ls, lseqm, lseqmdl), the source IMERG / CPC files, the version of the framework, and the run timestamp.

Trained model artefacts

trained_models/
  {prefix}_cnn_month{MM}_dekad{D}.keras
  {prefix}_cnn_month{MM}_dekad{D}_norm.json     # mean / std for input normalization

Loading: tf.keras.models.load_model(path). The matching _norm.json is required for applying the model to new data.

Confidence mask

station_density/confidence_mask_station_density.nc4

Variable: confidence (float32, 0-1). Computed once per AOI; cached and re-used across all dekads.

Metrics and QA

For each method:

metrics_{method}/
  {prefix}_metrics_{method}_month{MM}_dekad{D}.nc4    # 31 WMO metrics per pixel
quality_{method}/
  {prefix}_quality_{method}_month{MM}_dekad{D}.nc4    # CQI + 3 components

The metrics NetCDF holds the full WMO/TD-1485 + WMO-1317 set: RMSE, MAE, ME, Pearson r, NSE, KGE, POD, FAR, CSI, percentile errors, etc. The quality NetCDF is the aggregated CQI for plotting.

Station validation

station_validation/
  {prefix}_validation_by_station_{method}.csv     # one row per station, per method
  {prefix}_validation_by_region_{method}.csv      # aggregated by region_mapping
  {prefix}_taylor_by_station_{method}.png         # Taylor diagram
  {prefix}_taylor_monthly_{method}.png            # monthly Taylor by region

Figures

figures/
  {prefix}_<plot_type>_month{MM}_dekad{D}.png

Plot types include spatial comparison strips, CQI maps, station scatter, and Taylor diagrams. The visualisation hub notebook (06_visualisation_hub.ipynb) generates these.

--- title: "Output Structure" description: "Where everything lands and what each file contains." --- The output tree is organised by correction stage and by purpose. Every NetCDF carries CF-1.8 attributes - use `ncdump -h <file>` or `xr.open_dataset(<file>).attrs` to inspect. ## Top-level layout ``` data/output/ (or data/example_bali/output/ for Bali) corrected_ls/ corrected_lseqm/ corrected_lseqmdl/ trained_models/ station_density/ metrics_ls/ metrics_lseqm/ metrics_lseqmdl/ quality_ls/ quality_lseqm/ quality_lseqmdl/ station_validation/ figures/ ``` ## Corrected precipitation NetCDFs File pattern: ``` {prefix}_{method_abbr}_corrected_imergl_month{MM}_dekad{D}.nc4 ``` | Variable | Type | Units | Description | |----------|------|-------|-------------| | `precipitation` | float32 | mm/day | Corrected daily precipitation | | `lat`, `lon` | float64 | deg | Cell-centre coordinates, 0.1 deg grid | | `time` | datetime64 | UTC | Daily timestamps for the dekad x all years | Global attributes record the method (`ls`, `lseqm`, `lseqmdl`), the source IMERG / CPC files, the version of the framework, and the run timestamp. ## Trained model artefacts ``` trained_models/ {prefix}_cnn_month{MM}_dekad{D}.keras {prefix}_cnn_month{MM}_dekad{D}_norm.json # mean / std for input normalization ``` Loading: `tf.keras.models.load_model(path)`. The matching `_norm.json` is required for applying the model to new data. ## Confidence mask ``` station_density/confidence_mask_station_density.nc4 ``` Variable: `confidence` (float32, 0-1). Computed once per AOI; cached and re-used across all dekads. ## Metrics and QA For each method: ``` metrics_{method}/ {prefix}_metrics_{method}_month{MM}_dekad{D}.nc4 # 31 WMO metrics per pixel quality_{method}/ {prefix}_quality_{method}_month{MM}_dekad{D}.nc4 # CQI + 3 components ``` The metrics NetCDF holds the full WMO/TD-1485 + WMO-1317 set: RMSE, MAE, ME, Pearson r, NSE, KGE, POD, FAR, CSI, percentile errors, etc. The quality NetCDF is the aggregated CQI for plotting. ## Station validation ``` station_validation/ {prefix}_validation_by_station_{method}.csv # one row per station, per method {prefix}_validation_by_region_{method}.csv # aggregated by region_mapping {prefix}_taylor_by_station_{method}.png # Taylor diagram {prefix}_taylor_monthly_{method}.png # monthly Taylor by region ``` ## Figures ``` figures/ {prefix}_<plot_type>_month{MM}_dekad{D}.png ``` Plot types include spatial comparison strips, CQI maps, station scatter, and Taylor diagrams. The visualisation hub notebook (`06_visualisation_hub.ipynb`) generates these.