FIX: Min Max computation for physiographic descriptor

This commits resolves an issue when computing min and max value
of a physiographic descriptor when there are missing values that
are converted to -99

doc/source/release/1.0.0-notes.rst

@@ -705,3 +705,9 @@ Error in projected gradient value with ANN mapping
 
 There was an error in the values of the projected gradient when using ANN mapping. The current iteration was
 printing the projected gradient of the next iteration.
+
+Error when computing min and max of physiograhic descriptor
+***********************************************************
+
+There was an error when computing the min and max of a physiograhic descriptor if there is no data (i.e., 
+sea in the domain). No data from tif file are converted to -99 which leads to a wrong minimal value.

smash/core/model/_read_input_data.py

@@ -152,7 +152,6 @@ def _read_qobs(setup: SetupDT, mesh: MeshDT, input_data: Input_DataDT):
                     stacklevel=2,
                 )
             else:
-
                 ind_start_dat = max(0, start_diff)
                 ind_end_dat = min(dat.index.max(), end_diff)
                 ind_start_arr = max(0, -start_diff)
@@ -461,15 +460,28 @@ def _read_descriptor(setup: SetupDT, mesh: MeshDT, input_data: Input_DataDT):
             miss.append(name)
 
         else:
-            input_data.physio_data.descriptor[..., i] = _read_windowed_raster(path[0], mesh)
-            input_data.physio_data.l_descriptor[i] = np.min(input_data.physio_data.descriptor[..., i])
-            input_data.physio_data.u_descriptor[i] = np.max(input_data.physio_data.descriptor[..., i])
-            # % Check if descriptors are uniform
-            if input_data.physio_data.l_descriptor[i] == input_data.physio_data.u_descriptor[i]:
+            desc = _read_windowed_raster(path[0], mesh)
+            mask = desc != -99.0
+
+            # % Check if descriptor contains only no data
+            if not np.any(mask):
+                raise ValueError(
+                    f"Invalid descriptor '{name}'. It contains only missing values on the selected domain"
+                )
+
+            low = np.min(desc, where=mask, initial=np.inf)
+            upp = np.max(desc, where=mask, initial=-np.inf)
+
+            # % Check if descriptor is uniform
+            if low == upp:
                 raise ValueError(
-                    f"Reading spatially uniform descriptor '{name}'. It must be removed to perform "
-                    f"optimization"
+                    f"Invalid descriptor '{name}'. Spatially uniform values on the selected domain. It must "
+                    "be removed to perform optimization"
                 )
+            # % Assign values
+            input_data.physio_data.descriptor[..., i] = desc
+            input_data.physio_data.l_descriptor[i] = low
+            input_data.physio_data.u_descriptor[i] = upp
 
     if miss:
         warnings.warn(f"Missing {len(miss)} descriptor file(s): {miss}", stacklevel=2)