diff --git a/hat/data.py b/hat/data.py
index cfa04ed..b54abc3 100644
--- a/hat/data.py
+++ b/hat/data.py
@@ -218,25 +218,6 @@ def is_csv(file_path):
     return file_extension.lower() == ".csv"
 
 
-def read_csv_and_cache(fpath: str) -> gpd.GeoDataFrame:
-    """read .csv file and cache to pickle"""
-
-    # cache filepath
-    cache_fname = os.path.splitext(os.path.basename(fpath))[0]
-    cache_fpath = get_tmp_filepath(cache_fname, extension=".pickle")
-
-    # use cache if it exists
-    if os.path.exists(cache_fpath):
-        gdf = pd.read_pickle(cache_fpath)
-
-    # otherwise load from user defined filepath (and then cache)
-    else:
-        gdf = gpd.read_file(fpath)
-        gdf.to_pickle(cache_fpath)
-
-    return gdf
-
-
 """ other data (e.g. non geospatial)"""
 
 
diff --git a/hat/observations.py b/hat/observations.py
index 467eb70..a744f4a 100644
--- a/hat/observations.py
+++ b/hat/observations.py
@@ -1,7 +1,7 @@
 import geopandas as gpd
 import pandas as pd
 
-from hat.data import is_csv, read_csv_and_cache
+from hat.data import is_csv
 from hat.filters import filter_dataframe
 
 
@@ -28,7 +28,7 @@ def add_geometry_column(gdf: gpd.GeoDataFrame, coord_names):
     # Create a geometry column
     gdf["geometry"] = gpd.points_from_xy(gdf[x_coord_name], gdf[y_coord_name])
 
-    return gdf
+    return gpd.GeoDataFrame(gdf, geometry="geometry")
 
 
 def read_station_metadata_file(
@@ -39,7 +39,7 @@ def read_station_metadata_file(
 
     try:
         if is_csv(fpath):
-            gdf = read_csv_and_cache(fpath)
+            gdf = gpd.read_file(fpath)
             gdf = add_geometry_column(gdf, coord_names)
             gdf = gdf.set_crs(epsg=epsg)
         else: