diff --git a/src/parcels/_core/model.py b/src/parcels/_core/model.py
index 7a0945573..79c9c6460 100644
--- a/src/parcels/_core/model.py
+++ b/src/parcels/_core/model.py
@@ -103,12 +103,10 @@ def to_windowed_arrays(self, *, max_levels: int | None = None) -> Self:
             and ``max_levels`` then bounds its size.
         """
         windowed = self.__dict__.setdefault("_windowed", {})
-        for dim in ["lon", "lat"]:
-            # ensure lon and lat are loaded into memory for dask-backed datasets
-            if is_dask_collection(self.data[dim]):
+        for dim in ["lon", "lat", "depth"]:
+            # ensure lon, lat, depth are loaded into memory for dask-backed datasets
+            if dim in self.data and is_dask_collection(self.data[dim]):
                 self.data[dim].load()
-        if "depth" in self.data.dims and is_dask_collection(self.data["depth"]):
-            self.data["depth"].load()
         for name in self.scalar_field_names:
             current = windowed.get(name, self.data[name])
             windowed[name] = maybe_windowed(current, max_levels=max_levels)