Merge pull request #159 from GeoscienceAustralia/set_nodata_nan

Ensure floating point rasters have nodata set to `NaN`

Author: vnewey

intertidal/io.py

@@ -1059,9 +1059,12 @@ def _prepare_band(band, custom_dtypes, float_dtype, output_location, overwrite):
             band = band.fillna(int_nodata).astype(int_dtype)
             band.attrs["nodata"] = int_nodata
 
-        # Export other bands as float32 data types
+        # Export other bands as float32 data types, ensuring
+        # that nodata is explicitly set to np.nan in the outputs
+        # due to a missing nodata issue observed in 2026 annual runs
         else:
             band = band.astype(float_dtype)
+            band.attrs["nodata"] = np.nan
 
         # Export band to file
         if output_location is not None:

tests/README.md

@@ -10,7 +10,7 @@ Integration tests
 
 This directory contains tests that are run to verify that DEA Intertidal code runs correctly. The ``test_intertidal.py`` file runs a small-scale full workflow analysis over an intertidal flat in the Gulf of Carpentaria using the DEA Intertidal [Command Line Interface (CLI) tools](../notebooks/Intertidal_CLI.ipynb), and compares these results against a LiDAR validation DEM to produce some simple accuracy metrics.
 
-The latest integration test completed at **2026-03-19 14:33**. Compared to the previous run, it had an:
+The latest integration test completed at **2026-03-23 17:31**. Compared to the previous run, it had an:
 - RMSE accuracy of **0.14 m ( :heavy_minus_sign: no change)**
 - MAE accuracy of **0.12 m ( :heavy_minus_sign: no change)**
 - Bias of **0.12 m ( :heavy_minus_sign: no change)**

tests/validation.csv

@@ -142,3 +142,4 @@ time,Correlation,RMSE,MAE,R-squared,Bias,Regression slope
 2026-03-17 03:12:10.521642+00:00,0.975,0.145,0.122,0.95,0.116,1.121
 2026-03-19 02:11:55.124933+00:00,0.975,0.145,0.122,0.95,0.116,1.121
 2026-03-19 03:33:01.026986+00:00,0.975,0.145,0.122,0.95,0.116,1.121
+2026-03-23 06:31:50.184064+00:00,0.975,0.145,0.122,0.95,0.116,1.121